From 0931212bc692177cfc0ebcf016bc869dd4f88fd8 Mon Sep 17 00:00:00 2001 From: Zachary Sloan Date: Tue, 27 Nov 2012 14:44:14 -0600 Subject: Renamed webqtlDataSet.py to data_set.py Renamed the class webqtlDataset to DataSet Finished cisLRS and transLRS search types in d_search.py Fixed parent/f1 issue in show_trait.py --- wqflask/base/data_set.py | 162 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 162 insertions(+) create mode 100755 wqflask/base/data_set.py (limited to 'wqflask/base/data_set.py') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py new file mode 100755 index 00000000..992c673e --- /dev/null +++ b/wqflask/base/data_set.py @@ -0,0 +1,162 @@ +# Copyright (C) University of Tennessee Health Science Center, Memphis, TN. +# +# This program is free software: you can redistribute it and/or modify it +# under the terms of the GNU Affero General Public License +# as published by the Free Software Foundation, either version 3 of the +# License, or (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# See the GNU Affero General Public License for more details. +# +# This program is available from Source Forge: at GeneNetwork Project +# (sourceforge.net/projects/genenetwork/). +# +# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010) +# at rwilliams@uthsc.edu and xzhou15@uthsc.edu +# +# +# +# This module is used by GeneNetwork project (www.genenetwork.org) +# +# Created by GeneNetwork Core Team 2010/08/10 +# +# Last updated by GeneNetwork Core Team 2010/10/20 + +from htmlgen import HTMLgen2 as HT + +import webqtlConfig + + + +class DataSet(object): + """ + Dataset class defines a dataset in webqtl, can be either Microarray, + Published phenotype, genotype, or user input dataset(temp) + + """ + + def __init__(self, dbName, cursor=None): + + assert dbName + self.id = 0 + self.name = '' + self.type = '' + self.group = '' + self.cursor = cursor + + #temporary storage + if dbName.find('Temp') >= 0: + self.searchfield = ['name','description'] + self.disfield = ['name','description'] + self.type = 'Temp' + self.id = 1 + self.fullname = 'Temporary Storage' + self.shortname = 'Temp' + elif dbName.find('Publish') >= 0: + pass + elif dbName.find('Geno') >= 0: + self.searchfield = ['name','chr'] + self.disfield = ['name','chr','mb', 'source2', 'sequence'] + self.type = 'Geno' + else: #ProbeSet + self.searchfield = ['name','description','probe_target_description', + 'symbol','alias','genbankid','unigeneid','omim', + 'refseq_transcriptid','probe_set_specificity', 'probe_set_blat_score'] + self.disfield = ['name','symbol','description','probe_target_description', + 'chr','mb','alias','geneid','genbankid', 'unigeneid', 'omim', + 'refseq_transcriptid','blatseq','targetseq','chipid', 'comments', + 'strand_probe','strand_gene','probe_set_target_region', + 'probe_set_specificity', 'probe_set_blat_score','probe_set_blat_mb_start', + 'probe_set_blat_mb_end', 'probe_set_strand', + 'probe_set_note_by_rw', 'flag'] + self.type = 'ProbeSet' + self.name = dbName + if self.cursor and self.id == 0: + self.retrieveName() + + + # Delete this eventually + @property + def riset(): + Weve_Renamed_This_As_Group + + + def get_group(self): + assert self.cursor + if self.type == 'Publish': + query = ''' + SELECT + InbredSet.Name, InbredSet.Id + FROM + InbredSet, PublishFreeze + WHERE + PublishFreeze.InbredSetId = InbredSet.Id AND + PublishFreeze.Name = "%s" + ''' % self.name + elif self.type == 'Geno': + query = ''' + SELECT + InbredSet.Name, InbredSet.Id + FROM + InbredSet, GenoFreeze + WHERE + GenoFreeze.InbredSetId = InbredSet.Id AND + GenoFreeze.Name = "%s" + ''' % self.name + elif self.type == 'ProbeSet': + query = ''' + SELECT + InbredSet.Name, InbredSet.Id + FROM + InbredSet, ProbeSetFreeze, ProbeFreeze + WHERE + ProbeFreeze.InbredSetId = InbredSet.Id AND + ProbeFreeze.Id = ProbeSetFreeze.ProbeFreezeId AND + ProbeSetFreeze.Name = "%s" + ''' % self.name + else: + return "" + self.cursor.execute(query) + group, RIID = self.cursor.fetchone() + if group == 'BXD300': + group = "BXD" + self.group = group + self.group_id = RIID + return group + + + def retrieveName(self): + assert self.id == 0 and self.cursor + query = ''' + SELECT + Id, Name, FullName, ShortName + FROM + %sFreeze + WHERE + public > %d AND + (Name = "%s" OR FullName = "%s" OR ShortName = "%s") + '''% (self.type, webqtlConfig.PUBLICTHRESH, self.name, self.name, self.name) + try: + self.cursor.execute(query) + self.id,self.name,self.fullname,self.shortname=self.cursor.fetchone() + except: + raise KeyError, `self.name`+' doesn\'t exist.' + + + def genHTML(self, Class='c0dd'): + return HT.Href(text = HT.Span('%s Database' % self.fullname, Class= "fwb " + Class), + url= webqtlConfig.INFOPAGEHREF % self.name,target="_blank") + +class PhenotypeDataSet(DataSet): + + def __init__(self): + self.searchfield = ['name','post_publication_description','abstract','title','authors'] + self.disfield = ['name','pubmed_id', + 'pre_publication_description', 'post_publication_description', 'original_description', + 'pre_publication_abbreviation', 'post_publication_abbreviation', + 'lab_code', 'submitter', 'owner', 'authorized_users', + 'authors','title','abstract', 'journal','volume','pages','month', + 'year','sequence', 'units', 'comments'] + self.type = 'Publish' \ No newline at end of file -- cgit v1.2.3 From 94dd9844fb55f4576d3a079e9d5e59ebbf911b8c Mon Sep 17 00:00:00 2001 From: Zachary Sloan Date: Tue, 27 Nov 2012 17:59:17 -0600 Subject: Created subclass for each main data set type and moved the code for getting trait info that was in search_results.py into its respective class Renamed webqtlDataset to DataSet/create_dataset in webqtlTrait.py, webqtlDatabaseFunction.py, and CorrelationPage.py Got search page running again for mRNA assay data sets with these changes --- web/webqtl/search/SearchResultPage.py | 6 +- wqflask/base/data_set.py | 556 ++++++++++++++++++++----- wqflask/base/webqtlTrait.py | 25 +- wqflask/dbFunction/webqtlDatabaseFunction.py | 23 +- wqflask/wqflask/correlation/CorrelationPage.py | 4 +- wqflask/wqflask/do_search.py | 22 +- wqflask/wqflask/search_results.py | 325 +-------------- 7 files changed, 514 insertions(+), 447 deletions(-) (limited to 'wqflask/base/data_set.py') diff --git a/web/webqtl/search/SearchResultPage.py b/web/webqtl/search/SearchResultPage.py index 029a54c4..d62bb449 100755 --- a/web/webqtl/search/SearchResultPage.py +++ b/web/webqtl/search/SearchResultPage.py @@ -14,7 +14,7 @@ from htmlgen import HTMLgen2 as HT from base import webqtlConfig from utility.THCell import THCell from utility.TDCell import TDCell -from base.webqtlDataset import webqtlDataset +from base.data_set import DataSet from base.webqtlTrait import webqtlTrait from base.templatePage import templatePage from utility import webqtlUtil @@ -65,12 +65,12 @@ class SearchResultPage(templatePage): InbredSet where PublishFreeze.Name not like 'BXD300%' and InbredSet.Id = PublishFreeze.InbredSetId""") results = self.cursor.fetchall() - self.database = map(lambda x: webqtlDataset(x[0], self.cursor), results) + self.database = map(lambda x: DataSet(x[0], self.cursor), results) self.databaseCrosses = map(lambda x: x[1], results) self.databaseCrossIds = map(lambda x: x[2], results) self.singleCross = False else: - self.database = map(lambda x: webqtlDataset(x, self.cursor), self.database) + self.database = map(lambda x: DataSet(x, self.cursor), self.database) #currently, webqtl wouldn't allow multiple crosses #for other than multiple publish db search #so we can use the first database as example diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 992c673e..9e3e6d81 100755 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -19,64 +19,64 @@ # # # This module is used by GeneNetwork project (www.genenetwork.org) -# -# Created by GeneNetwork Core Team 2010/08/10 -# -# Last updated by GeneNetwork Core Team 2010/10/20 + +from __future__ import print_function, division from htmlgen import HTMLgen2 as HT import webqtlConfig +from pprint import pformat as pf +# Used by create_database to instantiate objects +DS_NAME_MAP = {} + +def create_dataset(db_conn, dataset_name): + cursor = db_conn.cursor() + cursor.execute(""" + SELECT DBType.Name + FROM DBList, DBType + WHERE DBList.Name = %s and + DBType.Id = DBList.DBTypeId + """, (dataset_name)) + print("dataset_name:", dataset_name) + dataset_type = cursor.fetchone()[0] + print("dataset_type:", pf(dataset_type)) + + dataset_ob = DS_NAME_MAP[dataset_type] + #dataset_class = getattr(data_set, dataset_ob) + + print("DS_NAME_MAP:", pf(DS_NAME_MAP)) + + dataset_class = globals()[dataset_ob] + return dataset_class(dataset_name, db_conn) class DataSet(object): """ - Dataset class defines a dataset in webqtl, can be either Microarray, + DataSet class defines a dataset in webqtl, can be either Microarray, Published phenotype, genotype, or user input dataset(temp) """ - def __init__(self, dbName, cursor=None): - - assert dbName - self.id = 0 - self.name = '' - self.type = '' - self.group = '' - self.cursor = cursor - - #temporary storage - if dbName.find('Temp') >= 0: - self.searchfield = ['name','description'] - self.disfield = ['name','description'] - self.type = 'Temp' - self.id = 1 - self.fullname = 'Temporary Storage' - self.shortname = 'Temp' - elif dbName.find('Publish') >= 0: - pass - elif dbName.find('Geno') >= 0: - self.searchfield = ['name','chr'] - self.disfield = ['name','chr','mb', 'source2', 'sequence'] - self.type = 'Geno' - else: #ProbeSet - self.searchfield = ['name','description','probe_target_description', - 'symbol','alias','genbankid','unigeneid','omim', - 'refseq_transcriptid','probe_set_specificity', 'probe_set_blat_score'] - self.disfield = ['name','symbol','description','probe_target_description', - 'chr','mb','alias','geneid','genbankid', 'unigeneid', 'omim', - 'refseq_transcriptid','blatseq','targetseq','chipid', 'comments', - 'strand_probe','strand_gene','probe_set_target_region', - 'probe_set_specificity', 'probe_set_blat_score','probe_set_blat_mb_start', - 'probe_set_blat_mb_end', 'probe_set_strand', - 'probe_set_note_by_rw', 'flag'] - self.type = 'ProbeSet' - self.name = dbName - if self.cursor and self.id == 0: - self.retrieveName() - - + def __init__(self, name, db_conn): + + assert name + self.name = name + self.db_conn = db_conn + self.cursor = self.db_conn.cursor() + self.id = None + self.type = None + self.group = None + + #if self.cursor and self.id == 0: + self.setup() + + self.check_confidentiality() + + self.retrieve_name() + self.get_group() + + # Delete this eventually @property def riset(): @@ -85,8 +85,93 @@ class DataSet(object): def get_group(self): assert self.cursor - if self.type == 'Publish': - query = ''' + self.cursor.execute(self.query) + self.group, self.group_id = self.cursor.fetchone() + if self.group == 'BXD300': + self.group = "BXD" + #return group + + + def retrieve_name(self): + """ + If the data set name parameter is not found in the 'Name' field of the data set table, + check if it is actually the FullName or ShortName instead. + + This is not meant to retrieve the data set info if no name at all is passed. + + """ + + query_args = tuple(self.db_conn.escape_string(x) for x in ( + (self.type + "Freeze"), + str(webqtlConfig.PUBLICTHRESH), + self.name, + self.name, + self.name)) + print("query_args are:", query_args) + + query = ''' + SELECT + Id, Name, FullName, ShortName + FROM + %s + WHERE + public > %s AND + (Name = "%s" OR FullName = "%s" OR ShortName = "%s") + ''' % (query_args) + + self.cursor.execute(query) + self.id, self.name, self.fullname, self.shortname = self.cursor.fetchone() + + + #def genHTML(self, Class='c0dd'): + # return HT.Href(text = HT.Span('%s Database' % self.fullname, Class= "fwb " + Class), + # url= webqtlConfig.INFOPAGEHREF % self.name,target="_blank") + +class PhenotypeDataSet(DataSet): + DS_NAME_MAP['Publish'] = 'PhenotypeDataSet' + + def setup(self): + # Fields in the database table + self.search_fields = ['Phenotype.Post_publication_description', + 'Phenotype.Pre_publication_description', + 'Phenotype.Pre_publication_abbreviation', + 'Phenotype.Post_publication_abbreviation', + 'Phenotype.Lab_code', + 'Publication.PubMed_ID', + 'Publication.Abstract', + 'Publication.Title', + 'Publication.Authors', + 'PublishXRef.Id'] + + # Figure out what display_fields is + self.display_fields = ['name', + 'pubmed_id', + 'pre_publication_description', + 'post_publication_description', + 'original_description', + 'pre_publication_abbreviation', + 'post_publication_abbreviation', + 'lab_code', + 'submitter', 'owner', + 'authorized_users', + 'authors', 'title', + 'abstract', 'journal', + 'volume', 'pages', + 'month', 'year', + 'sequence', 'units', 'comments'] + + # Fields displayed in the search results table header + self.header_fields = ['', + 'ID', + 'Description', + 'Authors', + 'Year', + 'Max LRS', + 'Max LRS Location'] + + self.type = 'Publish' + + self.query = ''' SELECT InbredSet.Name, InbredSet.Id FROM @@ -94,69 +179,336 @@ class DataSet(object): WHERE PublishFreeze.InbredSetId = InbredSet.Id AND PublishFreeze.Name = "%s" - ''' % self.name - elif self.type == 'Geno': - query = ''' - SELECT - InbredSet.Name, InbredSet.Id - FROM - InbredSet, GenoFreeze - WHERE - GenoFreeze.InbredSetId = InbredSet.Id AND - GenoFreeze.Name = "%s" - ''' % self.name - elif self.type == 'ProbeSet': - query = ''' - SELECT - InbredSet.Name, InbredSet.Id - FROM - InbredSet, ProbeSetFreeze, ProbeFreeze - WHERE - ProbeFreeze.InbredSetId = InbredSet.Id AND - ProbeFreeze.Id = ProbeSetFreeze.ProbeFreezeId AND - ProbeSetFreeze.Name = "%s" - ''' % self.name - else: - return "" - self.cursor.execute(query) - group, RIID = self.cursor.fetchone() - if group == 'BXD300': - group = "BXD" - self.group = group - self.group_id = RIID - return group + ''' % self.db_conn.escape_string(self.name) + + def check_confidentiality(self): + # (Urgently?) Need to write this + pass + + def get_trait_info(self, trait_list, species = ''): + for this_trait in trait_list: + if not this_trait.haveinfo: + this_trait.retrieveInfo(QTL=1) + + description = this_trait.post_publication_description + if this_trait.confidential: + if not webqtlUtil.hasAccessToConfidentialPhenotypeTrait(privilege=self.privilege, userName=self.userName, authorized_users=this_trait.authorized_users): + description = this_trait.pre_publication_description + this_trait.description_display = description + + if not this_trait.year.isdigit(): + this_trait.pubmed_text = "N/A" + + if this_trait.pubmed_id: + this_trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % this_trait.pubmed_id + + #LRS and its location + this_trait.LRS_score_repr = "N/A" + this_trait.LRS_score_value = 0 + this_trait.LRS_location_repr = "N/A" + this_trait.LRS_location_value = 1000000 + + if this_trait.lrs: + self.cursor.execute(""" + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '%s' and + Geno.Name = '%s' and + Geno.SpeciesId = Species.Id + """ % (species, this_trait.locus)) + result = self.cursor.fetchone() + if result: + if result[0] and result[1]: + LRS_Chr = result[0] + LRS_Mb = result[1] - def retrieveName(self): - assert self.id == 0 and self.cursor + #XZ: LRS_location_value is used for sorting + try: + LRS_location_value = int(LRS_Chr)*1000 + float(LRS_Mb) + except: + if LRS_Chr.upper() == 'X': + LRS_location_value = 20*1000 + float(LRS_Mb) + else: + LRS_location_value = ord(str(LRS_chr).upper()[0])*1000 + float(LRS_Mb) + + this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs + this_trait.LRS_score_value = LRS_score_value = this_trait.lrs + this_trait.LRS_location_repr = LRS_location_repr = 'Chr %s: %.4f Mb' % (LRS_Chr, float(LRS_Mb) ) + +class GenotypeDataSet(DataSet): + DS_NAME_MAP['Geno'] = 'GenotypeDataSet' + + def setup(self): + # Fields in the database table + self.search_fields = ['Name', + 'Chr'] + + # Find out what display_fields is + self.display_fields = ['name', + 'chr', + 'mb', + 'source2', + 'sequence'] + + # Fields displayed in the search results table header + self.header_fields = ['', + 'ID', + 'Location'] + + # Todo: Obsolete or rename this field + self.type = 'Geno' + query = ''' SELECT - Id, Name, FullName, ShortName + InbredSet.Name, InbredSet.Id FROM - %sFreeze + InbredSet, GenoFreeze WHERE - public > %d AND - (Name = "%s" OR FullName = "%s" OR ShortName = "%s") - '''% (self.type, webqtlConfig.PUBLICTHRESH, self.name, self.name, self.name) - try: + GenoFreeze.InbredSetId = InbredSet.Id AND + GenoFreeze.Name = "%s" + ''' % self.db_conn.escape_string(self.name) + + def check_confidentiality(self): + return geno_mrna_confidentiality(self) + + def get_trait_info(self, trait_list): + for this_trait in trait_list: + if not this_trait.haveinfo: + this_trait.retrieveInfo() + + #XZ: trait_location_value is used for sorting + trait_location_repr = 'N/A' + trait_location_value = 1000000 + + if this_trait.chr and this_trait.mb: + try: + trait_location_value = int(this_trait.chr)*1000 + this_trait.mb + except: + if this_trait.chr.upper() == 'X': + trait_location_value = 20*1000 + this_trait.mb + else: + trait_location_value = ord(str(this_trait.chr).upper()[0])*1000 + this_trait.mb + + this_trait.location_repr = 'Chr%s: %.4f' % (this_trait.chr, float(this_trait.mb) ) + this_trait.location_value = trait_location_value + + +class MrnaAssayDataSet(DataSet): + ''' + An mRNA Assay is a quantitative assessment (assay) associated with an mRNA trait + + This used to be called ProbeSet, but that term only refers specifically to the Affymetrix + platform and is far too specific. + + ''' + DS_NAME_MAP['ProbeSet'] = 'MrnaAssayDataSet' + + def setup(self): + # Fields in the database table + self.search_fields = ['Name', + 'Description', + 'Probe_Target_Description', + 'Symbol', + 'Alias', + 'GenbankId', + 'UniGeneId', + 'RefSeq_TranscriptId'] + + # Find out what display_fields is + self.display_fields = ['name', 'symbol', + 'description', 'probe_target_description', + 'chr', 'mb', + 'alias', 'geneid', + 'genbankid', 'unigeneid', + 'omim', 'refseq_transcriptid', + 'blatseq', 'targetseq', + 'chipid', 'comments', + 'strand_probe', 'strand_gene', + 'probe_set_target_region', + 'probe_set_specificity', + 'probe_set_blat_score', + 'probe_set_blat_mb_start', + 'probe_set_blat_mb_end', + 'probe_set_strand', + 'probe_set_note_by_rw', + 'flag'] + + # Fields displayed in the search results table header + self.header_fields = ['', + 'ID', + 'Symbol', + 'Description', + 'Location', + 'Mean Expr', + 'Max LRS', + 'Max LRS Location'] + + # Todo: Obsolete or rename this field + self.type = 'ProbeSet' + + self.query = ''' + SELECT + InbredSet.Name, InbredSet.Id + FROM + InbredSet, ProbeSetFreeze, ProbeFreeze + WHERE + ProbeFreeze.InbredSetId = InbredSet.Id AND + ProbeFreeze.Id = ProbeSetFreeze.ProbeFreezeId AND + ProbeSetFreeze.Name = "%s" + ''' % self.db_conn.escape_string(self.name) + + + def check_confidentiality(self): + return geno_mrna_confidentiality(self) + + def get_trait_info(self, trait_list=None, species=''): + + # Note: setting trait_list to [] is probably not a great idea. + if not trait_list: + trait_list = [] + + for this_trait in trait_list: + + if not this_trait.haveinfo: + this_trait.retrieveInfo(QTL=1) + + if this_trait.symbol: + pass + else: + this_trait.symbol = "N/A" + + #XZ, 12/08/2008: description + #XZ, 06/05/2009: Rob asked to add probe target description + description_string = str(this_trait.description).strip() + target_string = str(this_trait.probe_target_description).strip() + + description_display = '' + + if len(description_string) > 1 and description_string != 'None': + description_display = description_string + else: + description_display = this_trait.symbol + + if len(description_display) > 1 and description_display != 'N/A' and len(target_string) > 1 and target_string != 'None': + description_display = description_display + '; ' + target_string.strip() + + # Save it for the jinja2 tablet + this_trait.description_display = description_display + + #XZ: trait_location_value is used for sorting + trait_location_repr = 'N/A' + trait_location_value = 1000000 + + if this_trait.chr and this_trait.mb: + try: + trait_location_value = int(this_trait.chr)*1000 + this_trait.mb + except: + if this_trait.chr.upper() == 'X': + trait_location_value = 20*1000 + this_trait.mb + else: + trait_location_value = ord(str(this_trait.chr).upper()[0])*1000 + this_trait.mb + + this_trait.location_repr = 'Chr %s: %.4f Mb' % (this_trait.chr, float(this_trait.mb) ) + this_trait.location_value = trait_location_value + #this_trait.trait_location_value = trait_location_value + + #XZ, 01/12/08: This SQL query is much faster. + query = ( +"""select ProbeSetXRef.mean from ProbeSetXRef, ProbeSet + where ProbeSetXRef.ProbeSetFreezeId = %s and + ProbeSet.Id = ProbeSetXRef.ProbeSetId and + ProbeSet.Name = '%s' + """ % (self.db_conn.escape_string(str(this_trait.db.id)), + self.db_conn.escape_string(this_trait.name))) + + print("query is:", pf(query)) + self.cursor.execute(query) - self.id,self.name,self.fullname,self.shortname=self.cursor.fetchone() - except: - raise KeyError, `self.name`+' doesn\'t exist.' + result = self.cursor.fetchone() + if result: + if result[0]: + mean = result[0] + else: + mean=0 + else: + mean = 0 - def genHTML(self, Class='c0dd'): - return HT.Href(text = HT.Span('%s Database' % self.fullname, Class= "fwb " + Class), - url= webqtlConfig.INFOPAGEHREF % self.name,target="_blank") + #XZ, 06/05/2009: It is neccessary to turn on nowrap + this_trait.mean = repr = "%2.3f" % mean -class PhenotypeDataSet(DataSet): + #LRS and its location + this_trait.LRS_score_repr = 'N/A' + this_trait.LRS_score_value = 0 + this_trait.LRS_location_repr = 'N/A' + this_trait.LRS_location_value = 1000000 + + #Max LRS and its Locus location + if this_trait.lrs and this_trait.locus: + self.cursor.execute(""" + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '%s' and + Geno.Name = '%s' and + Geno.SpeciesId = Species.Id + """ % (species, this_trait.locus)) + result = self.cursor.fetchone() + + if result: + if result[0] and result[1]: + LRS_Chr = result[0] + LRS_Mb = result[1] + + #XZ: LRS_location_value is used for sorting + try: + LRS_location_value = int(LRS_Chr)*1000 + float(LRS_Mb) + except: + if LRS_Chr.upper() == 'X': + LRS_location_value = 20*1000 + float(LRS_Mb) + else: + LRS_location_value = ord(str(LRS_chr).upper()[0])*1000 + float(LRS_Mb) + + this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs + this_trait.LRS_score_value = LRS_score_value = this_trait.lrs + this_trait.LRS_location_repr = LRS_location_repr = 'Chr %s: %.4f Mb' % (LRS_Chr, float(LRS_Mb) ) + + +class TempDataSet(DataSet): + '''Temporary user-generated data set''' - def __init__(self): - self.searchfield = ['name','post_publication_description','abstract','title','authors'] - self.disfield = ['name','pubmed_id', - 'pre_publication_description', 'post_publication_description', 'original_description', - 'pre_publication_abbreviation', 'post_publication_abbreviation', - 'lab_code', 'submitter', 'owner', 'authorized_users', - 'authors','title','abstract', 'journal','volume','pages','month', - 'year','sequence', 'units', 'comments'] - self.type = 'Publish' \ No newline at end of file + def setup(self): + self.search_fields = ['name', + 'description'] + + self.display_fields = ['name', + 'description'] + + self.header_fields = ['Name', + 'Description'] + + self.type = 'Temp' + + # Need to double check later how these are used + self.id = 1 + self.fullname = 'Temporary Storage' + self.shortname = 'Temp' + + +def geno_mrna_confidentiality(ob): + dataset_table = ob.type + "Freeze" + print("dataset_table [%s]: %s" % (type(dataset_table), dataset_table)) + + query = '''SELECT Id, Name, FullName, confidentiality, + AuthorisedUsers FROM %s WHERE Name = %%s''' % (dataset_table) + + ob.cursor.execute(query, ob.name) + + (dataset_id, + name, + full_name, + confidential, + authorized_users) = ob.cursor.fetchall()[0] + + if confidential: + # Allow confidential data later + NoConfindetialDataForYouTodaySorry + \ No newline at end of file diff --git a/wqflask/base/webqtlTrait.py b/wqflask/base/webqtlTrait.py index 51d36ab2..29087721 100755 --- a/wqflask/base/webqtlTrait.py +++ b/wqflask/base/webqtlTrait.py @@ -6,7 +6,7 @@ from htmlgen import HTMLgen2 as HT import webqtlConfig from webqtlCaseData import webqtlCaseData -from webqtlDataset import webqtlDataset +from data_set import create_dataset from dbFunction import webqtlDatabaseFunction from utility import webqtlUtil @@ -20,9 +20,10 @@ class webqtlTrait: """ - def __init__(self, cursor = None, **kw): + def __init__(self, db_conn, **kw): print("in webqtlTrait") - self.cursor = cursor + self.db_conn = db_conn + self.cursor = self.db_conn.cursor() self.db = None # database object self.name = '' # Trait ID, ProbeSet ID, Published ID, etc. self.cellid = '' @@ -50,7 +51,7 @@ class webqtlTrait: if self.db and isinstance(self.db, basestring): assert self.cursor, "Don't have a cursor" - self.db = webqtlDataset(self.db, self.cursor) + self.db = create_dataset(self.db_conn, self.db) #if self.db == None, not from a database print("self.db is:", self.db, type(self.db)) @@ -396,8 +397,8 @@ class webqtlTrait: #XZ, 05/08/2009: Xiaodong add this block to use ProbeSet.Id to find the probeset instead of just using ProbeSet.Name #XZ, 05/08/2009: to avoid the problem of same probeset name from different platforms. elif self.db.type == 'ProbeSet': - disfieldString = string.join(self.db.disfield,',ProbeSet.') - disfieldString = 'ProbeSet.' + disfieldString + display_fields_string = ',ProbeSet.'.join(self.db.display_fields) + display_fields_string = 'ProbeSet.' + display_fields_string query = """ SELECT %s FROM ProbeSet, ProbeSetFreeze, ProbeSetXRef @@ -406,12 +407,12 @@ class webqtlTrait: ProbeSetXRef.ProbeSetId = ProbeSet.Id AND ProbeSetFreeze.Name = '%s' AND ProbeSet.Name = '%s' - """ % (disfieldString, self.db.name, self.name) + """ % (display_fields_string, self.db.name, self.name) #XZ, 05/08/2009: We also should use Geno.Id to find marker instead of just using Geno.Name # to avoid the problem of same marker name from different species. elif self.db.type == 'Geno': - disfieldString = string.join(self.db.disfield,',Geno.') - disfieldString = 'Geno.' + disfieldString + display_fields_string = string.join(self.db.display_fields,',Geno.') + display_fields_string = 'Geno.' + display_fields_string query = """ SELECT %s FROM Geno, GenoFreeze, GenoXRef @@ -420,10 +421,10 @@ class webqtlTrait: GenoXRef.GenoId = Geno.Id AND GenoFreeze.Name = '%s' AND Geno.Name = '%s' - """ % (disfieldString, self.db.name, self.name) + """ % (display_fields_string, self.db.name, self.name) else: #Temp type query = 'SELECT %s FROM %s WHERE Name = "%s"' % \ - (string.join(self.db.disfield,','), self.db.type, self.name) + (string.join(self.db.display_fields,','), self.db.type, self.name) self.cursor.execute(query) @@ -432,7 +433,7 @@ class webqtlTrait: self.haveinfo = 1 #XZ: assign SQL query result to trait attributes. - for i, field in enumerate(self.db.disfield): + for i, field in enumerate(self.db.display_fields): setattr(self, field, traitInfo[i]) if self.db.type == 'Publish': diff --git a/wqflask/dbFunction/webqtlDatabaseFunction.py b/wqflask/dbFunction/webqtlDatabaseFunction.py index 7e33da3f..8f923b8a 100755 --- a/wqflask/dbFunction/webqtlDatabaseFunction.py +++ b/wqflask/dbFunction/webqtlDatabaseFunction.py @@ -19,14 +19,7 @@ # # # This module is used by GeneNetwork project (www.genenetwork.org) -# -# Created by GeneNetwork Core Team 2010/08/10 -# -# Last updated by Xiaodong Zhou 2011/Jan/20 -#webqtlDatabaseFunction.py -# -#This file consists of various database related functions; the names are generally self-explanatory. import MySQLdb import string @@ -206,21 +199,21 @@ def getTissueCountByTissueProbeSetFreezeId(cursor=None, TissueProbeSetFreezeId=N ########################################################################### # input: cursor, TissueProbeSetFreezeId (int) -# output: DatasetName(string),DatasetFullName(string) -# function: retrieve DatasetName, DatasetFullName based on TissueProbeSetFreezeId +# output: DataSetName(string),DataSetFullName(string) +# function: retrieve DataSetName, DataSetFullName based on TissueProbeSetFreezeId ########################################################################### -def getDatasetNamesByTissueProbeSetFreezeId(cursor=None, TissueProbeSetFreezeId=None): +def getDataSetNamesByTissueProbeSetFreezeId(cursor=None, TissueProbeSetFreezeId=None): query ="select Name, FullName from TissueProbeSetFreeze where Id=%s" % TissueProbeSetFreezeId try: cursor.execute(query) result = cursor.fetchone() - DatasetName = result[0] - DatasetFullName =result[1] + DataSetName = result[0] + DataSetFullName =result[1] except: - DatasetName =None - DatasetFullName =None + DataSetName =None + DataSetFullName =None - return DatasetName, DatasetFullName + return DataSetName, DataSetFullName ########################################################################### # input: cursor, geneIdLst (list) diff --git a/wqflask/wqflask/correlation/CorrelationPage.py b/wqflask/wqflask/correlation/CorrelationPage.py index e48ea412..8af30d1e 100644 --- a/wqflask/wqflask/correlation/CorrelationPage.py +++ b/wqflask/wqflask/correlation/CorrelationPage.py @@ -47,7 +47,7 @@ from base import webqtlConfig from utility.THCell import THCell from utility.TDCell import TDCell from base.webqtlTrait import webqtlTrait -from base.webqtlDataset import webqtlDataset +from base.data_set import create_dataset from base.templatePage import templatePage from utility import webqtlUtil from dbFunction import webqtlDatabaseFunction @@ -310,7 +310,7 @@ class CorrelationPage(templatePage): #try: #print("target_db_name is:", target_db_name) - self.db = webqtlDataset(self.target_db_name, self.cursor) + self.db = create_dataset(self.db_conn, self.target_db_name) #except: # detail = ["The database you just requested has not been established yet."] # self.error(detail) diff --git a/wqflask/wqflask/do_search.py b/wqflask/wqflask/do_search.py index e2bafb3a..73a72e00 100644 --- a/wqflask/wqflask/do_search.py +++ b/wqflask/wqflask/do_search.py @@ -147,7 +147,7 @@ class GenotypeSearch(DoSearch): """WHERE %s and Geno.Id = GenoXRef.GenoId and GenoXRef.GenoFreezeId = GenoFreeze.Id and - GenoFreeze.Id = %s"""% ( + GenoFreeze.Id = %s""" % ( self.get_where_clause(), self.escape(self.dataset.id))) @@ -257,7 +257,7 @@ class GoSearch(ProbeSetSearch): statements = ("""%s.symbol=GOgene_product.symbol and GOassociation.gene_product_id=GOgene_product.id and GOterm.id=GOassociation.term_id""" % ( - self.db_conn.escape_string(self.dataset.type))) + self.escape(self.dataset.type))) where_clause = " %s = '%s' and %s " % (field, go_id, statements) @@ -317,14 +317,14 @@ class CisLrsSearch(ProbeSetSearch): Geno.SpeciesId = %s and %s.Chr = Geno.Chr and ABS(%s.Mb-Geno.Mb) < %s """ % ( - self.dataset.type, + self.escape(self.dataset.type), min(lower_limit, upper_limit), - self.dataset.type, + self.escape(self.dataset.type), max(lower_limit, upper_limit), - self.dataset.type, + self.escape(self.dataset.type), self.species_id, - self.dataset.type, - self.dataset.type, + self.escape(self.dataset.type), + self.escape(self.dataset.type), min_threshold ) else: @@ -437,7 +437,7 @@ if __name__ == "__main__": from base import webqtlConfig - from base.webqtlDataset import webqtlDataset + from base.data_set import create_dataset from base.templatePage import templatePage from utility import webqtlUtil from dbFunction import webqtlDatabaseFunction @@ -449,13 +449,13 @@ if __name__ == "__main__": cursor = db_conn.cursor() dataset_name = "HC_M2_0606_P" - dataset = webqtlDataset(dataset_name, cursor) + dataset = create_dataset(db_conn, dataset_name) #results = ProbeSetSearch("salt", dataset, cursor, db_conn).run() #results = RifSearch("diabetes", dataset, cursor, db_conn).run() #results = WikiSearch("nicotine", dataset, cursor, db_conn).run() - results = TransLrsSearch(['25','99','10'], dataset, cursor, db_conn).run() - #results = TransLrsSearch(['9', '999', '10'], dataset, cursor, db_conn).run() + results = CisLrsSearch(['25','99','10'], dataset, cursor, db_conn).run() + #results = TransLrsSearch(['25', '999', '10'], dataset, cursor, db_conn).run() #results = PhenotypeSearch("brain", dataset, cursor, db_conn).run() #results = GenotypeSearch("rs13475699", dataset, cursor, db_conn).run() #results = GoSearch("0045202", dataset, cursor, db_conn).run() diff --git a/wqflask/wqflask/search_results.py b/wqflask/wqflask/search_results.py index 05f062fc..b50e45d5 100644 --- a/wqflask/wqflask/search_results.py +++ b/wqflask/wqflask/search_results.py @@ -10,7 +10,7 @@ from flask import render_template # # ################################################### -import string +#import string import os import cPickle import re @@ -29,7 +29,7 @@ from htmlgen import HTMLgen2 as HT from base import webqtlConfig from utility.THCell import THCell from utility.TDCell import TDCell -from base.webqtlDataset import webqtlDataset +from base.data_set import create_dataset from base.webqtlTrait import webqtlTrait from base.templatePage import templatePage from wqflask import parser @@ -43,14 +43,13 @@ from utility import formatting class SearchResultPage(templatePage): + #maxReturn = 3000 - maxReturn = 3000 - nkeywords = 0 def __init__(self, fd): print("initing SearchResultPage") - import logging_tree - logging_tree.printout() + #import logging_tree + #logging_tree.printout() self.fd = fd templatePage.__init__(self, fd) assert self.openMysql(), "Couldn't open MySQL" @@ -59,127 +58,40 @@ class SearchResultPage(templatePage): self.dataset = fd['dataset'] # change back to self.dataset - if not self.dataset or self.dataset == 'spacer': - #Error, No dataset selected - heading = "Search Result" - detail = ['''No dataset was selected for this search, please - go back and SELECT at least one dataset.'''] - self.error(heading=heading,detail=detail,error="No dataset Selected") - return + #if not self.dataset or self.dataset == 'spacer': + # #Error, No dataset selected + # heading = "Search Result" + # detail = ['''No dataset was selected for this search, please + # go back and SELECT at least one dataset.'''] + # self.error(heading=heading,detail=detail,error="No dataset Selected") + # return ########################################### # Names and IDs of RISet / F2 set ########################################### + + # All Phenotypes is a special case we'll deal with later if self.dataset == "All Phenotypes": self.cursor.execute(""" select PublishFreeze.Name, InbredSet.Name, InbredSet.Id from PublishFreeze, InbredSet where PublishFreeze.Name not like 'BXD300%' and InbredSet.Id = PublishFreeze.InbredSetId""") results = self.cursor.fetchall() - self.dataset = map(lambda x: webqtlDataset(x[0], self.cursor), results) + self.dataset = map(lambda x: DataSet(x[0], self.cursor), results) self.dataset_groups = map(lambda x: x[1], results) self.dataset_group_ids = map(lambda x: x[2], results) - self.single_group = False else: print("self.dataset is:", pf(self.dataset)) - self.dataset = webqtlDataset(self.dataset, self.cursor) + self.dataset = create_dataset(self.db_conn, self.dataset) print("self.dataset is now:", pf(self.dataset)) - if self.dataset.type in ("Geno", "ProbeSet"): - db_type = self.dataset.type + "Freeze" - print("db_type [%s]: %s" % (type(db_type), db_type)) - - query = '''SELECT Id, Name, FullName, confidentiality, - AuthorisedUsers FROM %s WHERE Name = %%s''' % (db_type) - - self.cursor.execute(query, self.dataset.name) - - (indId, - indName, - indFullName, - confidential, - AuthorisedUsers) = self.cursor.fetchall()[0] - - if confidential: - # Allow confidential data later - NoConfindetialDataForYouTodaySorry - #access_to_confidential_dataset = 0 - # - ##for the dataset that confidentiality is 1 - ##1. 'admin' and 'root' can see all of the dataset - ##2. 'user' can see the dataset that AuthorisedUsers contains his id(stored in the Id field of User table) - #if webqtlConfig.USERDICT[self.privilege] > webqtlConfig.USERDICT['user']: - # access_to_confidential_dataset = 1 - #else: - # AuthorisedUsersList=AuthorisedUsers.split(',') - # if AuthorisedUsersList.__contains__(self.userName): - # access_to_confidential_dataset = 1 - # - #if not access_to_confidential_dataset: - # Some error - - #else: - # heading = "Search Result" - # detail = ['''The dataset has not been established yet, please - # go back and SELECT at least one dataset.'''] - # self.error(heading=heading,detail=detail,error="No dataset Selected") - # return - - self.dataset.get_group() - self.single_group = True - #XZ, August 24,2010: Since self.single_group = True, it's safe to assign one species Id. - self.species_id = webqtlDatabaseFunction.retrieveSpeciesId(self.cursor, - self.dataset.group) - - #self.db_type = self.dataset.type - if self.dataset.type == "Publish": - self.search_fields = ['Phenotype.Post_publication_description', - 'Phenotype.Pre_publication_description', - 'Phenotype.Pre_publication_abbreviation', - 'Phenotype.Post_publication_abbreviation', - 'Phenotype.Lab_code', - 'Publication.PubMed_ID', - 'Publication.Abstract', - 'Publication.Title', - 'Publication.Authors', - 'PublishXRef.Id'] - self.header_fields = ['', - 'ID', - 'Description', - 'Authors', - 'Year', - 'Max LRS', - 'Max LRS Location'] - - elif self.dataset.type == "ProbeSet": - self.search_fields = ['Name', - 'Description', - 'Probe_Target_Description', - 'Symbol', - 'Alias', - 'GenbankId', - 'UniGeneId', - 'RefSeq_TranscriptId'] - self.header_fields = ['', - 'ID', - 'Symbol', - 'Description', - 'Location', - 'Mean Expr', - 'Max LRS', - 'Max LRS Location'] - elif self.dataset.type == "Geno": - self.search_fields = ['Name', - 'Chr'] - self.header_fields = ['', - 'ID', - 'Location'] - + self.search() self.gen_search_result() def gen_search_result(self): - """Get the info displayed in the search result table from the set of results computed in + """ + Get the info displayed in the search result table from the set of results computed in the "search" function """ @@ -191,26 +103,19 @@ class SearchResultPage(templatePage): if not result: continue - seq = 1 group = self.dataset.group - species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, RISet=group) #### Excel file needs to be generated #### print("foo locals are:", locals()) trait_id = result[0] - this_trait = webqtlTrait(db=self.dataset, name=trait_id, cursor=self.cursor) + this_trait = webqtlTrait(self.db_conn, db=self.dataset, name=trait_id) this_trait.retrieveInfo(QTL=True) print("this_trait is:", pf(this_trait)) self.trait_list.append(this_trait) - - if self.dataset.type == "ProbeSet": - self.getTraitInfoForProbeSet(trait_list=self.trait_list, species=species) - elif self.dataset.type == "Publish": - self.getTraitInfoForPublish(trait_list=self.trait_list, species=species) - elif self.dataset.type == "Geno": - self.getTraitInfoForGeno(trait_list=self.trait_list) + + self.dataset.get_trait_info(self.trait_list, species) def search(self): @@ -222,7 +127,7 @@ class SearchResultPage(templatePage): print("[kodak] item is:", pf(a_search)) search_term = a_search['search_term'] if a_search['key']: - search_type = string.upper(a_search['key']) + search_type = a_search['key'].upper() else: # We fall back to the dataset type as the key to get the right object search_type = self.dataset.type @@ -258,187 +163,3 @@ class SearchResultPage(templatePage): keyword = string.replace(keyword,"?",".") wildcardkeyword[i] = keyword#'[[:<:]]'+ keyword+'[[:>:]]' return wildcardkeyword - - - def getTraitInfoForGeno(self, trait_list): - for this_trait in trait_list: - if not this_trait.haveinfo: - this_trait.retrieveInfo() - - #XZ: trait_location_value is used for sorting - trait_location_repr = 'N/A' - trait_location_value = 1000000 - - if this_trait.chr and this_trait.mb: - try: - trait_location_value = int(this_trait.chr)*1000 + this_trait.mb - except: - if this_trait.chr.upper() == 'X': - trait_location_value = 20*1000 + this_trait.mb - else: - trait_location_value = ord(str(this_trait.chr).upper()[0])*1000 + this_trait.mb - - this_trait.location_repr = 'Chr%s: %.4f' % (this_trait.chr, float(this_trait.mb) ) - this_trait.location_value = trait_location_value - - - def getTraitInfoForPublish(self, trait_list, species = ''): - for this_trait in trait_list: - if not this_trait.haveinfo: - this_trait.retrieveInfo(QTL=1) - - description = this_trait.post_publication_description - if this_trait.confidential: - if not webqtlUtil.hasAccessToConfidentialPhenotypeTrait(privilege=self.privilege, userName=self.userName, authorized_users=this_trait.authorized_users): - description = this_trait.pre_publication_description - this_trait.description_display = description - - if not this_trait.year.isdigit(): - this_trait.pubmed_text = "N/A" - - if this_trait.pubmed_id: - this_trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % this_trait.pubmed_id - - #LRS and its location - this_trait.LRS_score_repr = "N/A" - this_trait.LRS_score_value = 0 - this_trait.LRS_location_repr = "N/A" - this_trait.LRS_location_value = 1000000 - - if this_trait.lrs: - self.cursor.execute(""" - select Geno.Chr, Geno.Mb from Geno, Species - where Species.Name = '%s' and - Geno.Name = '%s' and - Geno.SpeciesId = Species.Id - """ % (species, this_trait.locus)) - result = self.cursor.fetchone() - - if result: - if result[0] and result[1]: - LRS_Chr = result[0] - LRS_Mb = result[1] - - #XZ: LRS_location_value is used for sorting - try: - LRS_location_value = int(LRS_Chr)*1000 + float(LRS_Mb) - except: - if LRS_Chr.upper() == 'X': - LRS_location_value = 20*1000 + float(LRS_Mb) - else: - LRS_location_value = ord(str(LRS_chr).upper()[0])*1000 + float(LRS_Mb) - - this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_score_value = LRS_score_value = this_trait.lrs - this_trait.LRS_location_repr = LRS_location_repr = 'Chr %s: %.4f Mb' % (LRS_Chr, float(LRS_Mb) ) - - - def getTraitInfoForProbeSet(self, trait_list=None, species=''): - - # Note: setting trait_list to [] is probably not a great idea. - if not trait_list: - trait_list = [] - - for this_trait in trait_list: - - if not this_trait.haveinfo: - this_trait.retrieveInfo(QTL=1) - - if this_trait.symbol: - pass - else: - this_trait.symbol = "N/A" - - #XZ, 12/08/2008: description - #XZ, 06/05/2009: Rob asked to add probe target description - description_string = str(this_trait.description).strip() - target_string = str(this_trait.probe_target_description).strip() - - description_display = '' - - if len(description_string) > 1 and description_string != 'None': - description_display = description_string - else: - description_display = this_trait.symbol - - if len(description_display) > 1 and description_display != 'N/A' and len(target_string) > 1 and target_string != 'None': - description_display = description_display + '; ' + target_string.strip() - - # Save it for the jinja2 tablet - this_trait.description_display = description_display - - #XZ: trait_location_value is used for sorting - trait_location_repr = 'N/A' - trait_location_value = 1000000 - - if this_trait.chr and this_trait.mb: - try: - trait_location_value = int(this_trait.chr)*1000 + this_trait.mb - except: - if this_trait.chr.upper() == 'X': - trait_location_value = 20*1000 + this_trait.mb - else: - trait_location_value = ord(str(this_trait.chr).upper()[0])*1000 + this_trait.mb - - this_trait.location_repr = 'Chr %s: %.4f Mb' % (this_trait.chr, float(this_trait.mb) ) - this_trait.location_value = trait_location_value - #this_trait.trait_location_value = trait_location_value - - #XZ, 01/12/08: This SQL query is much faster. - query = ( -"""select ProbeSetXRef.mean from ProbeSetXRef, ProbeSet - where ProbeSetXRef.ProbeSetFreezeId = %s and - ProbeSet.Id = ProbeSetXRef.ProbeSetId and - ProbeSet.Name = '%s' - """ % (self.db_conn.escape_string(str(this_trait.db.id)), - self.db_conn.escape_string(this_trait.name))) - - print("query is:", pf(query)) - - self.cursor.execute(query) - result = self.cursor.fetchone() - - if result: - if result[0]: - mean = result[0] - else: - mean=0 - else: - mean = 0 - - #XZ, 06/05/2009: It is neccessary to turn on nowrap - this_trait.mean = repr = "%2.3f" % mean - - #LRS and its location - this_trait.LRS_score_repr = 'N/A' - this_trait.LRS_score_value = 0 - this_trait.LRS_location_repr = 'N/A' - this_trait.LRS_location_value = 1000000 - - #Max LRS and its Locus location - if this_trait.lrs and this_trait.locus: - self.cursor.execute(""" - select Geno.Chr, Geno.Mb from Geno, Species - where Species.Name = '%s' and - Geno.Name = '%s' and - Geno.SpeciesId = Species.Id - """ % (species, this_trait.locus)) - result = self.cursor.fetchone() - - if result: - if result[0] and result[1]: - LRS_Chr = result[0] - LRS_Mb = result[1] - - #XZ: LRS_location_value is used for sorting - try: - LRS_location_value = int(LRS_Chr)*1000 + float(LRS_Mb) - except: - if LRS_Chr.upper() == 'X': - LRS_location_value = 20*1000 + float(LRS_Mb) - else: - LRS_location_value = ord(str(LRS_chr).upper()[0])*1000 + float(LRS_Mb) - - this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_score_value = LRS_score_value = this_trait.lrs - this_trait.LRS_location_repr = LRS_location_repr = 'Chr %s: %.4f Mb' % (LRS_Chr, float(LRS_Mb) ) -- cgit v1.2.3 From d1f2863c15f62ae37833fa1311870d7b1aab3355 Mon Sep 17 00:00:00 2001 From: Zachary Sloan Date: Wed, 28 Nov 2012 14:30:04 -0600 Subject: Made some small changes to get code working for genotype searches --- wqflask/base/data_set.py | 2 +- wqflask/wqflask/parser.py | 2 ++ wqflask/wqflask/search_results.py | 5 +++-- 3 files changed, 6 insertions(+), 3 deletions(-) (limited to 'wqflask/base/data_set.py') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 9e3e6d81..d9d3a52b 100755 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -258,7 +258,7 @@ class GenotypeDataSet(DataSet): # Todo: Obsolete or rename this field self.type = 'Geno' - query = ''' + self.query = ''' SELECT InbredSet.Name, InbredSet.Id FROM diff --git a/wqflask/wqflask/parser.py b/wqflask/wqflask/parser.py index dc33fc52..7711942a 100644 --- a/wqflask/wqflask/parser.py +++ b/wqflask/wqflask/parser.py @@ -15,6 +15,8 @@ Both square brackets and parentheses can be used interchangeably. Both can also encapsulate a single value; "cisLRS=[9 999 10)" would be acceptable.] +NEED TO DEAL WITH WILDCARD CHARACTER '*' + """ from __future__ import print_function, division diff --git a/wqflask/wqflask/search_results.py b/wqflask/wqflask/search_results.py index b50e45d5..96350f22 100644 --- a/wqflask/wqflask/search_results.py +++ b/wqflask/wqflask/search_results.py @@ -55,6 +55,7 @@ class SearchResultPage(templatePage): assert self.openMysql(), "Couldn't open MySQL" print("fd is:", pf(fd)) + print("fd.dict is:", pf(fd['dataset'])) self.dataset = fd['dataset'] # change back to self.dataset @@ -93,7 +94,7 @@ class SearchResultPage(templatePage): """ Get the info displayed in the search result table from the set of results computed in the "search" function - + """ self.trait_list = [] # result_set represents the results for each search term; a search of @@ -114,7 +115,7 @@ class SearchResultPage(templatePage): this_trait.retrieveInfo(QTL=True) print("this_trait is:", pf(this_trait)) self.trait_list.append(this_trait) - + self.dataset.get_trait_info(self.trait_list, species) -- cgit v1.2.3 From 94300b4488aa334ced34981981ad5d0ecdec01d6 Mon Sep 17 00:00:00 2001 From: Zachary Sloan Date: Thu, 29 Nov 2012 18:44:01 -0600 Subject: Changed a number of variables (riset to group, db to dataset) Put most of the code for cisLRS and transLRS searches into the class CisTransLrsSearch (might change this name to something else later) Simplified escape code for searches in do_search.py Got search_results working again after some changes --- wqflask/base/data_set.py | 5 +- wqflask/base/webqtlFormData.py | 26 +-- wqflask/base/webqtlTrait.py | 146 ++++++------- wqflask/dbFunction/webqtlDatabaseFunction.py | 4 +- wqflask/wqflask/do_search.py | 237 +++++++++------------- wqflask/wqflask/search_results.py | 18 +- wqflask/wqflask/show_trait/show_trait.py | 83 ++++---- wqflask/wqflask/templates/index_page.html | 4 +- wqflask/wqflask/templates/search_result_page.html | 2 +- wqflask/wqflask/views.py | 3 + 10 files changed, 241 insertions(+), 287 deletions(-) (limited to 'wqflask/base/data_set.py') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index d9d3a52b..633f7545 100755 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -392,8 +392,9 @@ class MrnaAssayDataSet(DataSet): if len(description_display) > 1 and description_display != 'N/A' and len(target_string) > 1 and target_string != 'None': description_display = description_display + '; ' + target_string.strip() - # Save it for the jinja2 tablet + # Save it for the jinja2 template this_trait.description_display = description_display + #print(" xxxxdd [%s]: %s" % (type(this_trait.description_display), description_display)) #XZ: trait_location_value is used for sorting trait_location_repr = 'N/A' @@ -418,7 +419,7 @@ class MrnaAssayDataSet(DataSet): where ProbeSetXRef.ProbeSetFreezeId = %s and ProbeSet.Id = ProbeSetXRef.ProbeSetId and ProbeSet.Name = '%s' - """ % (self.db_conn.escape_string(str(this_trait.db.id)), + """ % (self.db_conn.escape_string(str(this_trait.dataset.id)), self.db_conn.escape_string(this_trait.name))) print("query is:", pf(query)) diff --git a/wqflask/base/webqtlFormData.py b/wqflask/base/webqtlFormData.py index ff1db0e8..a3537c87 100755 --- a/wqflask/base/webqtlFormData.py +++ b/wqflask/base/webqtlFormData.py @@ -47,7 +47,7 @@ from utility import webqtlUtil class webqtlFormData(object): 'Represents data from a WebQTL form page, needed to generate the next page' - attrs = ('formID','RISet','genotype','samplelist','allsamplelist', 'display_variance' + attrs = ('formID','group','genotype','samplelist','allsamplelist', 'display_variance' 'suggestive','significance','submitID','identification', 'enablevariance', 'nperm','nboot','email','incparentsf1','genotype_1','genotype_2','traitInfo') @@ -104,11 +104,11 @@ class webqtlFormData(object): self.ppolar = None self.mpolar = None - print("[yellow] self.RISet is:", self.RISet) - if self.RISet: + print("[yellow] self.group is:", self.group) + if self.group: #try: # # NL, 07/27/2010. ParInfo has been moved from webqtlForm.py to webqtlUtil.py; - _f1, _f12, self.mpolar, self.ppolar = webqtlUtil.ParInfo[self.RISet] + _f1, _f12, self.mpolar, self.ppolar = webqtlUtil.ParInfo[self.group] #except: # f1 = f12 = self.mpolar = self.ppolar = None @@ -129,8 +129,8 @@ class webqtlFormData(object): #self.readGenotype() #self.readData() - if self.RISet == 'BXD300': - self.RISet = 'BXD' + if self.group == 'BXD300': + self.group = 'BXD' def __getitem__(self, key): @@ -153,17 +153,17 @@ class webqtlFormData(object): def readGenotype(self): '''read genotype from .geno file''' - if self.RISet == 'BXD300': - self.RISet = 'BXD' + if self.group == 'BXD300': + self.group = 'BXD' - assert self.RISet, "self.RISet needs to be set" + assert self.group, "self.group needs to be set" #genotype_1 is Dataset Object without parents and f1 #genotype_2 is Dataset Object with parents and f1 (not for intercross) self.genotype_1 = reaper.Dataset() - full_filename = os.path.join(webqtlConfig.GENODIR, self.RISet + '.geno') + full_filename = os.path.join(webqtlConfig.GENODIR, self.group + '.geno') # reaper barfs on unicode filenames, so here we ensure it's a string full_filename = str(full_filename) @@ -173,12 +173,12 @@ class webqtlFormData(object): try: # NL, 07/27/2010. ParInfo has been moved from webqtlForm.py to webqtlUtil.py; - _f1, _f12, _mat, _pat = webqtlUtil.ParInfo[self.RISet] + _f1, _f12, _mat, _pat = webqtlUtil.ParInfo[self.group] except KeyError: _f1 = _f12 = _mat = _pat = None self.genotype_2 = self.genotype_1 - if self.genotype_1.type == "riset" and _mat and _pat: + if self.genotype_1.type == "group" and _mat and _pat: self.genotype_2 = self.genotype_1.add(Mat=_mat, Pat=_pat) #, F1=_f1) #determine default genotype object @@ -333,7 +333,7 @@ class webqtlFormData(object): def Sample(self): 'Create some dummy data for testing' - self.RISet = 'BXD' + self.group = 'BXD' self.incparentsf1 = 'on' #self.display = 9.2 #self.significance = 16.1 diff --git a/wqflask/base/webqtlTrait.py b/wqflask/base/webqtlTrait.py index 29087721..cc0e2321 100755 --- a/wqflask/base/webqtlTrait.py +++ b/wqflask/base/webqtlTrait.py @@ -24,11 +24,11 @@ class webqtlTrait: print("in webqtlTrait") self.db_conn = db_conn self.cursor = self.db_conn.cursor() - self.db = None # database object + self.dataset = None # database object self.name = '' # Trait ID, ProbeSet ID, Published ID, etc. self.cellid = '' self.identification = 'un-named trait' - self.riset = '' + self.group = '' self.haveinfo = 0 self.sequence = '' # Blat sequence, available for ProbeSet self.data = {} @@ -41,22 +41,22 @@ class webqtlTrait: elif name == 'fullname': name2 = value.split("::") if len(name2) == 2: - self.db, self.name = name2 + self.dataset, self.name = name2 elif len(name2) == 3: - self.db, self.name, self.cellid = name2 + self.dataset, self.name, self.cellid = name2 else: raise KeyError, repr(value) + ' parameter format error.' else: raise KeyError, repr(name) + ' not a valid parameter for this class.' - if self.db and isinstance(self.db, basestring): + if self.dataset and isinstance(self.dataset, basestring): assert self.cursor, "Don't have a cursor" - self.db = create_dataset(self.db_conn, self.db) + self.dataset = create_dataset(self.db_conn, self.dataset) - #if self.db == None, not from a database - print("self.db is:", self.db, type(self.db)) - if self.db: - if self.db.type == "Temp": + #if self.dataset == None, not from a database + print("self.dataset is:", self.dataset, type(self.dataset)) + if self.dataset: + if self.dataset.type == "Temp": self.cursor.execute(''' SELECT InbredSet.Name @@ -66,9 +66,11 @@ class webqtlTrait: Temp.InbredSetId = InbredSet.Id AND Temp.Name = "%s" ''', self.name) - self.riset = self.cursor.fetchone()[0] + self.group = self.cursor.fetchone()[0] else: - self.riset = self.db.get_group() + self.group = self.dataset.get_group() + + print("trinity, self.group is:", self.group) # # In ProbeSet, there are maybe several annotations match one sequence @@ -82,8 +84,8 @@ class webqtlTrait: # The variable self.sequence should be changed to self.BlatSeq # It also should be changed in other places where it are used. - if self.db: - if self.db.type == 'ProbeSet': + if self.dataset: + if self.dataset.type == 'ProbeSet': print("Doing ProbeSet Query") query = ''' SELECT @@ -95,7 +97,7 @@ class webqtlTrait: ProbeSetFreeze.Id = ProbeSetXRef.ProbeSetFreezeId and ProbeSet.Name = %s and ProbeSetFreeze.Name = %s - ''', (self.name, self.db.name) + ''', (self.name, self.dataset.name) print("query is:", query) self.cursor.execute(*query) self.sequence = self.cursor.fetchone()[0] @@ -104,8 +106,8 @@ class webqtlTrait: def getName(self): str = "" - if self.db and self.name: - str = "%s::%s" % (self.db, self.name) + if self.dataset and self.name: + str = "%s::%s" % (self.dataset, self.name) if self.cellid: str += "::" + self.cellid else: @@ -124,8 +126,8 @@ class webqtlTrait: # def getGivenName(self): str = self.name - if self.db and self.name: - if self.db.type=='Temp': + if self.dataset and self.name: + if self.dataset.type=='Temp': self.cursor.execute('SELECT description FROM Temp WHERE Name=%s', self.name) desc = self.cursor.fetchone()[0] if desc.__contains__('PCA'): @@ -137,16 +139,16 @@ class webqtlTrait: def displayName(self): str = "" - if self.db and self.name: - if self.db.type=='Temp': + if self.dataset and self.name: + if self.dataset.type=='Temp': desc = self.description if desc.__contains__('PCA'): desc = desc[desc.rindex(':')+1:].strip() else: desc = desc[:desc.index('entered')].strip() - str = "%s::%s" % (self.db, desc) + str = "%s::%s" % (self.dataset, desc) else: - str = "%s::%s" % (self.db, self.name) + str = "%s::%s" % (self.dataset, self.name) if self.cellid: str += "::" + self.cellid else: @@ -156,7 +158,7 @@ class webqtlTrait: #def __str__(self): - # #return "%s %s" % (self.getName(), self.riset) + # #return "%s %s" % (self.getName(), self.group) # return self.getName() #__str__ = getName #__repr__ = __str__ @@ -207,7 +209,7 @@ class webqtlTrait: # def getSequence(self): assert self.cursor - if self.db.type == 'ProbeSet': + if self.dataset.type == 'ProbeSet': self.cursor.execute(''' SELECT ProbeSet.BlatSeq @@ -218,7 +220,7 @@ class webqtlTrait: ProbeSetFreeze.Id = ProbeSetXRef.ProbSetFreezeId and ProbeSet.Name = %s ProbeSetFreeze.Name = %s - ''', self.name, self.db.name) + ''', self.name, self.dataset.name) #self.cursor.execute(query) results = self.fetchone() @@ -230,9 +232,9 @@ class webqtlTrait: if samplelist == None: samplelist = [] - assert self.db and self.cursor + assert self.dataset and self.cursor - if self.db.type == 'Temp': + if self.dataset.type == 'Temp': query = ''' SELECT Strain.Name, TempData.value, TempData.SE, TempData.NStrain, TempData.Id @@ -246,7 +248,7 @@ class webqtlTrait: Strain.Name ''' % self.name #XZ, 03/02/2009: Xiaodong changed Data to PublishData, SE to PublishSE - elif self.db.type == 'Publish': + elif self.dataset.type == 'Publish': query = ''' SELECT Strain.Name, PublishData.value, PublishSE.error, NStrain.count, PublishData.Id @@ -263,7 +265,7 @@ class webqtlTrait: PublishFreeze.Id = %d AND PublishData.StrainId = Strain.Id Order BY Strain.Name - ''' % (self.name, self.db.id) + ''' % (self.name, self.dataset.id) #XZ, 03/02/2009: Xiaodong changed Data to ProbeData, SE to ProbeSE elif self.cellid: @@ -287,9 +289,9 @@ class webqtlTrait: ProbeData.StrainId = Strain.Id Order BY Strain.Name - ''' % (self.cellid, self.name, self.db.name) + ''' % (self.cellid, self.name, self.dataset.name) #XZ, 03/02/2009: Xiaodong added this block for ProbeSetData and ProbeSetSE - elif self.db.type == 'ProbeSet': + elif self.dataset.type == 'ProbeSet': #ProbeSet Data query = ''' SELECT @@ -306,7 +308,7 @@ class webqtlTrait: ProbeSetData.StrainId = Strain.Id Order BY Strain.Name - ''' % (self.name, self.db.name) + ''' % (self.name, self.dataset.name) #XZ, 03/02/2009: Xiaodong changeded Data to GenoData, SE to GenoSE else: #Geno Data @@ -326,7 +328,7 @@ class webqtlTrait: GenoData.StrainId = Strain.Id Order BY Strain.Name - ''' % (webqtlDatabaseFunction.retrieveSpeciesId(self.cursor, self.db.riset), self.name, self.db.name) + ''' % (webqtlDatabaseFunction.retrieveSpeciesId(self.cursor, self.dataset.group), self.name, self.dataset.name) self.cursor.execute(query) @@ -341,7 +343,7 @@ class webqtlTrait: if not samplelist or (samplelist and name in samplelist): #if value != None: # num_cases = None - # if self.db.type in ('Publish', 'Temp'): + # if self.dataset.type in ('Publish', 'Temp'): # ndata = item[3] name = item[0] self.data[name] = webqtlCaseData(*item) #name, value, variance, num_cases) @@ -352,7 +354,7 @@ class webqtlTrait: # if val != None: # var = item[2] # ndata = None - # if self.db.type in ('Publish', 'Temp'): + # if self.dataset.type in ('Publish', 'Temp'): # ndata = item[3] # self.data[item[0]] = webqtlCaseData(val, var, ndata) # #end for @@ -370,9 +372,9 @@ class webqtlTrait: # return self.__dict__.items() def retrieveInfo(self, QTL = None): - assert self.db and self.cursor - if self.db.type == 'Publish': - #self.db.DisField = ['Name','PubMed_ID','Phenotype','Abbreviation','Authors','Title',\ + assert self.dataset and self.cursor + if self.dataset.type == 'Publish': + #self.dataset.DisField = ['Name','PubMed_ID','Phenotype','Abbreviation','Authors','Title',\ # 'Abstract', 'Journal','Volume','Pages','Month','Year','Sequence',\ # 'Units', 'comments'] query = ''' @@ -393,11 +395,11 @@ class webqtlTrait: Publication.Id = PublishXRef.PublicationId AND PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND PublishFreeze.Id =%s - ''' % (self.name, self.db.id) + ''' % (self.name, self.dataset.id) #XZ, 05/08/2009: Xiaodong add this block to use ProbeSet.Id to find the probeset instead of just using ProbeSet.Name #XZ, 05/08/2009: to avoid the problem of same probeset name from different platforms. - elif self.db.type == 'ProbeSet': - display_fields_string = ',ProbeSet.'.join(self.db.display_fields) + elif self.dataset.type == 'ProbeSet': + display_fields_string = ',ProbeSet.'.join(self.dataset.display_fields) display_fields_string = 'ProbeSet.' + display_fields_string query = """ SELECT %s @@ -407,11 +409,11 @@ class webqtlTrait: ProbeSetXRef.ProbeSetId = ProbeSet.Id AND ProbeSetFreeze.Name = '%s' AND ProbeSet.Name = '%s' - """ % (display_fields_string, self.db.name, self.name) + """ % (display_fields_string, self.dataset.name, self.name) #XZ, 05/08/2009: We also should use Geno.Id to find marker instead of just using Geno.Name # to avoid the problem of same marker name from different species. - elif self.db.type == 'Geno': - display_fields_string = string.join(self.db.display_fields,',Geno.') + elif self.dataset.type == 'Geno': + display_fields_string = string.join(self.dataset.display_fields,',Geno.') display_fields_string = 'Geno.' + display_fields_string query = """ SELECT %s @@ -421,10 +423,10 @@ class webqtlTrait: GenoXRef.GenoId = Geno.Id AND GenoFreeze.Name = '%s' AND Geno.Name = '%s' - """ % (display_fields_string, self.db.name, self.name) + """ % (display_fields_string, self.dataset.name, self.name) else: #Temp type query = 'SELECT %s FROM %s WHERE Name = "%s"' % \ - (string.join(self.db.display_fields,','), self.db.type, self.name) + (string.join(self.dataset.display_fields,','), self.dataset.type, self.name) self.cursor.execute(query) @@ -433,16 +435,16 @@ class webqtlTrait: self.haveinfo = 1 #XZ: assign SQL query result to trait attributes. - for i, field in enumerate(self.db.display_fields): + for i, field in enumerate(self.dataset.display_fields): setattr(self, field, traitInfo[i]) - if self.db.type == 'Publish': + if self.dataset.type == 'Publish': self.confidential = 0 if self.pre_publication_description and not self.pubmed_id: self.confidential = 1 self.homologeneid = None - if self.db.type == 'ProbeSet' and self.riset and self.geneid: + if self.dataset.type == 'ProbeSet' and self.group and self.geneid: #XZ, 05/26/2010: From time to time, this query get error message because some geneid values in database are not number. #XZ: So I have to test if geneid is number before execute the query. #XZ: The geneid values in database should be cleaned up. @@ -463,7 +465,7 @@ class webqtlTrait: InbredSet.Name = '%s' AND InbredSet.SpeciesId = Species.Id AND Species.TaxonomyId = Homologene.TaxonomyId - """ % (self.geneid, self.riset) + """ % (self.geneid, self.group) self.cursor.execute(query) result = self.cursor.fetchone() else: @@ -473,7 +475,7 @@ class webqtlTrait: self.homologeneid = result[0] if QTL: - if self.db.type == 'ProbeSet' and not self.cellid: + if self.dataset.type == 'ProbeSet' and not self.cellid: query = ''' SELECT ProbeSetXRef.Locus, ProbeSetXRef.LRS, ProbeSetXRef.pValue, ProbeSetXRef.mean @@ -483,14 +485,14 @@ class webqtlTrait: ProbeSetXRef.ProbeSetId = ProbeSet.Id AND ProbeSet.Name = "%s" AND ProbeSetXRef.ProbeSetFreezeId =%s - ''' % (self.name, self.db.id) + ''' % (self.name, self.dataset.id) self.cursor.execute(query) traitQTL = self.cursor.fetchone() if traitQTL: self.locus, self.lrs, self.pvalue, self.mean = traitQTL else: self.locus = self.lrs = self.pvalue = self.mean = "" - if self.db.type == 'Publish': + if self.dataset.type == 'Publish': query = ''' SELECT PublishXRef.Locus, PublishXRef.LRS @@ -500,7 +502,7 @@ class webqtlTrait: PublishXRef.Id = %s AND PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND PublishFreeze.Id =%s - ''' % (self.name, self.db.id) + ''' % (self.name, self.dataset.id) self.cursor.execute(query) traitQTL = self.cursor.fetchone() if traitQTL: @@ -514,7 +516,7 @@ class webqtlTrait: if not self.haveinfo: self.retrieveInfo() - if self.db.type == 'Publish': + if self.dataset.type == 'Publish': PubMedLink = "" if self.pubmed_id: PubMedLink = HT.Href(text="PubMed %d : " % self.pubmed_id, @@ -524,10 +526,10 @@ class webqtlTrait: if formName: setDescription2 = HT.Href(url="javascript:showDatabase3('%s','%s','%s','')" % - (formName, self.db.name, self.name), Class = "fs14") + (formName, self.dataset.name, self.name), Class = "fs14") else: setDescription2 = HT.Href(url="javascript:showDatabase2('%s','%s','')" % - (self.db.name,self.name), Class = "fs14") + (self.dataset.name,self.name), Class = "fs14") if self.confidential and not webqtlUtil.hasAccessToConfidentialPhenotypeTrait(privilege=privilege, userName=userName, authorized_users=authorized_users): setDescription2.append('RecordID/%s - %s' % (self.name, self.pre_publication_description)) @@ -545,20 +547,20 @@ class webqtlTrait: setDescription2.append(HT.Italic('%s, and colleagues' % a1)) setDescription = HT.Span(PubMedLink, setDescription2) - elif self.db.type == 'Temp': + elif self.dataset.type == 'Temp': setDescription = HT.Href(text="%s" % (self.description),url="javascript:showDatabase2\ - ('%s','%s','')" % (self.db.name,self.name), Class = "fs14") + ('%s','%s','')" % (self.dataset.name,self.name), Class = "fs14") setDescription = HT.Span(setDescription) - elif self.db.type == 'Geno': # Genome DB only available for single search + elif self.dataset.type == 'Geno': # Genome DB only available for single search if formName: setDescription = HT.Href(text="Locus %s [Chr %s @ %s Mb]" % (self.name,self.chr,\ '%2.3f' % self.mb),url="javascript:showDatabase3('%s','%s','%s','')" % \ - (formName, self.db.name, self.name), Class = "fs14") + (formName, self.dataset.name, self.name), Class = "fs14") else: setDescription = HT.Href(text="Locus %s [Chr %s @ %s Mb]" % (self.name,self.chr,\ '%2.3f' % self.mb),url="javascript:showDatabase2('%s','%s','')" % \ - (self.db.name,self.name), Class = "fs14") + (self.dataset.name,self.name), Class = "fs14") setDescription = HT.Span(setDescription) @@ -566,20 +568,20 @@ class webqtlTrait: if self.cellid: if formName: setDescription = HT.Href(text="ProbeSet/%s/%s" % (self.name, self.cellid),url=\ - "javascript:showDatabase3('%s','%s','%s','%s')" % (formName, self.db.name,self.name,self.cellid), \ + "javascript:showDatabase3('%s','%s','%s','%s')" % (formName, self.dataset.name,self.name,self.cellid), \ Class = "fs14") else: setDescription = HT.Href(text="ProbeSet/%s/%s" % (self.name,self.cellid),url=\ - "javascript:showDatabase2('%s','%s','%s')" % (self.db.name,self.name,self.cellid), \ + "javascript:showDatabase2('%s','%s','%s')" % (self.dataset.name,self.name,self.cellid), \ Class = "fs14") else: if formName: setDescription = HT.Href(text="ProbeSet/%s" % self.name, url=\ - "javascript:showDatabase3('%s','%s','%s','')" % (formName, self.db.name,self.name), \ + "javascript:showDatabase3('%s','%s','%s','')" % (formName, self.dataset.name,self.name), \ Class = "fs14") else: setDescription = HT.Href(text="ProbeSet/%s" % self.name, url=\ - "javascript:showDatabase2('%s','%s','')" % (self.db.name,self.name), \ + "javascript:showDatabase2('%s','%s','')" % (self.dataset.name,self.name), \ Class = "fs14") if self.symbol and self.chr and self.mb: setDescription.append(' [') @@ -591,9 +593,9 @@ class webqtlTrait: setDescription.append('; %s' % self.probe_target_description) setDescription = HT.Span(setDescription) - if self.db.type != 'Temp' and dispFromDatabase: + if self.dataset.type != 'Temp' and dispFromDatabase: setDescription.append( ' --- FROM : ') - setDescription.append(self.db.genHTML(Class='cori')) + setDescription.append(self.dataset.genHTML(Class='cori')) return setDescription @property @@ -654,13 +656,13 @@ class webqtlTrait: select ProbeFreeze.Name from ProbeFreeze, ProbeSetFreeze where ProbeFreeze.Id = ProbeSetFreeze.ProbeFreezeId AND - ProbeSetFreeze.Id = %d""" % thisTrait.db.id) + ProbeSetFreeze.Id = %d""" % thisTrait.dataset.id) probeDBName = self.cursor.fetchone()[0] return dict(name = probeDBName, url = None) else: - return dict(name = self.db.fullname, - url = webqtlConfig.INFOPAGEHREF % self.db.name) + return dict(name = self.dataset.fullname, + url = webqtlConfig.INFOPAGEHREF % self.dataset.name) def calculate_correlation(self, values, method): """Calculate the correlation value and p value according to the method specified""" diff --git a/wqflask/dbFunction/webqtlDatabaseFunction.py b/wqflask/dbFunction/webqtlDatabaseFunction.py index 8f923b8a..1e028ecc 100755 --- a/wqflask/dbFunction/webqtlDatabaseFunction.py +++ b/wqflask/dbFunction/webqtlDatabaseFunction.py @@ -80,9 +80,9 @@ def getAllSpecies(cursor=None): #function: retrieve specie's name info based on RISet ########################################################################### -def retrieveSpecies(cursor=None, RISet=None): +def retrieveSpecies(cursor=None, group=None): try: - cursor.execute("select Species.Name from Species, InbredSet where InbredSet.Name = '%s' and InbredSet.SpeciesId = Species.Id" % RISet) + cursor.execute("select Species.Name from Species, InbredSet where InbredSet.Name = '%s' and InbredSet.SpeciesId = Species.Id" % group) return cursor.fetchone()[0] except: return None diff --git a/wqflask/wqflask/do_search.py b/wqflask/wqflask/do_search.py index 2b8efd68..92a754e3 100644 --- a/wqflask/wqflask/do_search.py +++ b/wqflask/wqflask/do_search.py @@ -20,7 +20,7 @@ class DoSearch(object): def __init__(self, search_term, search_operator, dataset, cursor, db_conn): self.search_term = search_term # Make sure search_operator is something we expect - assert search_operator in ("=", "<", ">", "<=", ">="), "Bad search operator" + assert search_operator in (None, "=", "<", ">", "<=", ">="), "Bad search operator" self.search_operator = search_operator self.dataset = dataset self.db_conn = db_conn @@ -41,6 +41,12 @@ class DoSearch(object): def escape(self, stringy): """Shorter name than self.db_conn.escape_string""" return self.db_conn.escape_string(str(stringy)) + + def mescape(self, *items): + """Multiple escape""" + escaped = [self.escape(item) for item in items] + print("escaped is:", escaped) + return tuple(escaped) def normalize_spaces(self, stringy): """Strips out newlines/extra spaces and replaces them with just spaces""" @@ -91,8 +97,7 @@ class ProbeSetSearch(DoSearch): """Generates and runs a simple search of an mRNA expression dataset""" print("Running ProbeSetSearch") - query = (self.base_query + - """WHERE (MATCH (ProbeSet.Name, + query = self.base_query + """WHERE (MATCH (ProbeSet.Name, ProbeSet.description, ProbeSet.symbol, alias, @@ -102,8 +107,8 @@ class ProbeSetSearch(DoSearch): AGAINST ('%s' IN BOOLEAN MODE)) and ProbeSet.Id = ProbeSetXRef.ProbeSetId and ProbeSetXRef.ProbeSetFreezeId = %s - """ % (self.escape(self.search_term), - self.escape(self.dataset.id))) + """ % (self.escape(self.search_term[0]), + self.escape(self.dataset.id)) print("final query is:", pf(query)) @@ -275,7 +280,8 @@ class GoSearch(ProbeSetSearch): class LrsSearch(ProbeSetSearch): """Searches for genes with a QTL within the given LRS values - LRS searches can take 2 different forms: + LRS searches can take 3 different forms: + - LRS > (or <) min/max_LRS - LRS=(min_LRS max_LRS) - LRS=(min_LRS max_LRS chromosome start_Mb end_Mb) where min/max_LRS represent the range of LRS scores and start/end_Mb represent @@ -289,129 +295,128 @@ class LrsSearch(ProbeSetSearch): self.search_term = [float(value) for value in self.search_term] - from_clause = ", Geno" + self.from_clause = ", Geno" if self.search_operator == "=": - if len(self.search_term) >= 2: - if len(self.search_term) == 2: - lrs_min, lrs_max = self.search_term - elif len(self.search_term) == 5: - lrs_min, lrs_max, chr_num, mb_low, mb_high = self.search_term - else: - SomeError - - sub_clause = """ %sXRef.LRS > %s and - %sXRef.LRS < %s and """ % (self.escape(self.dataset.type), - self.escape(min(lrs_min, lrs_max)), - self.escape(self.dataset.type), - self.escape(max(lrs_min, lrs_max))) - + assert isinstance(self.search_term, (list, tuple)) + self.lrs_min, self.lrs_max = self.search_term[:2] + + self.sub_clause = """ %sXRef.LRS > %s and + %sXRef.LRS < %s and """ % self.mescape(self.dataset.type, + min(self.lrs_min, self.lrs_max), + self.dataset.type, + max(self.lrs_min, self.lrs_max)) + + if len(self.search_term) > 2: + self.chr_num = self.search_term[2] + self.sub_clause += """ Geno.Chr = %s and """ % (self.escape(self.chr_num)) if len(self.search_term) == 5: - sub_clause = sub_clause + """ Geno.Mb > %s and + self.mb_low, self.mb_high = self.search_term[3:] + self.sub_clause += """ Geno.Mb > %s and Geno.Mb < %s and - Geno.Chr = %s and - """ % (self.escape(min(mb_low, mb_high)), - self.escape(max(mb_low, mb_high)), - self.escape(chr_num)) + """ % self.mescape(min(self.mb_low, self.mb_high), + max(self.mb_low, self.mb_high)) + print("self.sub_clause is:", pf(self.sub_clause)) else: # Deal with >, <, >=, and <= - sub_clause = """ %sXRef.LRS %s %s and """ % (self.escape(self.dataset.type), - self.escape(self.search_operator), - self.escape(self.search_term[0])) + self.sub_clause = """ %sXRef.LRS %s %s and """ % self.mescape(self.dataset.type, + self.search_operator, + self.search_term[0]) - where_clause = sub_clause + """ %sXRef.Locus = Geno.name and + self.where_clause = self.sub_clause + """ %sXRef.Locus = Geno.name and Geno.SpeciesId = %s and %s.Chr = Geno.Chr - """ % (self.escape(self.dataset.type), - self.escape(self.species_id), - self.escape(self.dataset.type)) + """ % self.mescape(self.dataset.type, + self.species_id, + self.dataset.type) - print("where_clause is:", pf(where_clause)) + print("where_clause is:", pf(self.where_clause)) - query = self.compile_final_query(from_clause, where_clause) + self.query = self.compile_final_query(self.from_clause, self.where_clause) - return self.execute(query) - -class CisLrsSearch(LrsSearch): - """Searches for genes on a particular chromosome with a cis-eQTL within the given LRS values + return self.execute(self.query) - A cisLRS search can take 3 forms: - - cisLRS=(min_LRS max_LRS) - - cisLRS=(min_LRS max_LRS mb_buffer) - - cisLRS>min_LRS - where min/max_LRS represent the range of LRS scores and the mb_buffer is the range around - a particular QTL where its eQTL would be considered "cis". If there is no third parameter, - mb_buffer will default to 5 megabases. - A QTL is a cis-eQTL if a gene's expression is regulated by a QTL in roughly the same area - (where the area is determined by the mb_buffer that the user can choose). +class CisTransLrsSearch(LrsSearch): - """ - - # This is tentatively a child of LrsSearch; I'll need to check what code, if any, overlaps - # between this and the LrsSearch code. In the original code, commands are divided by - # the number of inputs they take, so these commands are completely separate - - DoSearch.search_types['CISLRS'] = "CisLrsSearch" - - def run(self): + def real_run(self, the_operator): #if isinstance(self.search_term, basestring): # self.search_term = [self.search_term] print("self.search_term is:", self.search_term) self.search_term = [float(value) for value in self.search_term] - mb_buffer = 5 # default - - from_clause = ", Geno " - + self.mb_buffer = 5 # default + self.from_clause = ", Geno " + if self.search_operator == "=": if len(self.search_term) == 2: - lower_limit, upper_limit = self.search_term + self.lrs_min, self.lrs_max = self.search_term #[int(value) for value in self.search_term] elif len(self.search_term) == 3: - lower_limit, upper_limit, mb_buffer = self.search_term + self.lrs_min, self.lrs_max, self.mb_buffer = self.search_term else: SomeError - sub_clause = """ %sXRef.LRS > %s and - %sXRef.LRS < %s and - ABS(%s.Mb-Geno.Mb) < %s and """ % ( - self.escape(self.dataset.type), - self.escape(min(lower_limit, upper_limit)), + self.sub_clause = """ %sXRef.LRS > %s and + %sXRef.LRS < %s and """ % ( self.escape(self.dataset.type), - self.escape(max(lower_limit, upper_limit)), + self.escape(min(self.lrs_min, self.lrs_max)), self.escape(self.dataset.type), - self.escape(mb_buffer) + self.escape(max(self.lrs_min, self.lrs_max)) ) - else: # Deal with >, <, >=, and <= - sub_clause = """ %sXRef.LRS %s %s and - ABS(%s.Mb-Geno.Mb) < %s and """ % ( + self.sub_clause = """ %sXRef.LRS %s %s and """ % ( self.escape(self.dataset.type), self.escape(self.search_operator), - self.escape(self.search_term[0]), - self.escape(self.dataset.type), - self.escape(mb_buffer) + self.escape(self.search_term[0]) ) - - where_clause = sub_clause + """%sXRef.Locus = Geno.name and + + self.where_clause = self.sub_clause + """ + ABS(%s.Mb-Geno.Mb) %s %s and + %sXRef.Locus = Geno.name and Geno.SpeciesId = %s and %s.Chr = Geno.Chr""" % ( + self.escape(self.dataset.type), + the_operator, + self.escape(self.mb_buffer), self.escape(self.dataset.type), self.escape(self.species_id), self.escape(self.dataset.type) ) - print("where_clause is:", pf(where_clause)) + print("where_clause is:", pf(self.where_clause)) - query = self.compile_final_query(from_clause, where_clause) + self.query = self.compile_final_query(self.from_clause, self.where_clause) - return self.execute(query) + return self.execute(self.query) + + +class CisLrsSearch(CisTransLrsSearch): + """Searches for genes on a particular chromosome with a cis-eQTL within the given LRS values + + A cisLRS search can take 3 forms: + - cisLRS=(min_LRS max_LRS) + - cisLRS=(min_LRS max_LRS mb_buffer) + - cisLRS>min_LRS + where min/max_LRS represent the range of LRS scores and the mb_buffer is the range around + a particular QTL where its eQTL would be considered "cis". If there is no third parameter, + mb_buffer will default to 5 megabases. + + A QTL is a cis-eQTL if a gene's expression is regulated by a QTL in roughly the same area + (where the area is determined by the mb_buffer that the user can choose). + + """ + + DoSearch.search_types['CISLRS'] = "CisLrsSearch" + + def run(self): + return self.real_run("<") + -class TransLrsSearch(LrsSearch): +class TransLrsSearch(CisTransLrsSearch): """Searches for genes on a particular chromosome with a cis-eQTL within the given LRS values A transLRS search can take 2 forms: @@ -425,70 +430,11 @@ class TransLrsSearch(LrsSearch): (where the area is determined by the mb_buffer that the user can choose). Opposite of cis-eQTL. """ - - # This is tentatively a child of LrsSearch; I'll need to check what code, if any, overlaps - # between this and the LrsSearch code. In the original code, commands are divided by - # the number of inputs they take, so these commands are completely separate DoSearch.search_types['TRANSLRS'] = "TransLrsSearch" def run(self): - if len(self.search_term) == 3: - lower_limit, upper_limit, min_threshold = [int(value) for value in self.search_term] - - where_clause = """ %sXRef.LRS > %s and - %sXRef.LRS < %s and - %sXRef.Locus = Geno.name and - Geno.SpeciesId = %s and - (%s.Chr != Geno.Chr or - ABS(%s.Mb-Geno.Mb) > %s) """ % ( - self.dataset.type, - min(lower_limit, upper_limit), - self.dataset.type, - max(lower_limit, upper_limit), - self.dataset.type, - self.species_id, - self.dataset.type, - self.dataset.type, - min_threshold - ) - - else: - NeedSomeErrorHere - - return None - - -#itemCmd = item[0] -#lowerLimit = float(item[1]) -#upperLimit = float(item[2]) -# -#if itemCmd.upper() in ("TRANSLRS", "CISLRS"): -# if item[3]: -# mthresh = float(item[3]) -# clauseItem = " %sXRef.LRS > %2.7f and %sXRef.LRS < %2.7f " % \ -# (self.dbType, min(lowerLimit, upperLimit), self.dbType, max(lowerLimit, upperLimit)) -# if itemCmd.upper() == "CISLRS": -# clauseItem += """ and %sXRef.Locus = Geno.name and Geno.SpeciesId = %s and %s.Chr = Geno.Chr and ABS(%s.Mb-Geno.Mb) < %2.7f """ % (self.dbType, self.speciesId, self.dbType, self.dbType, mthresh) -# DescriptionText.append(HT.Span(' with a ', HT.U('cis-QTL'), ' having an LRS between %g and %g using a %g Mb exclusion buffer' % (min(lowerLimit, upperLimit), max(lowerLimit, upperLimit), mthresh))) -# else: -# clauseItem += """ and %sXRef.Locus = Geno.name and Geno.SpeciesId = %s and (%s.Chr != Geno.Chr or (%s.Chr != Geno.Chr and ABS(%s.Mb-Geno.Mb) > %2.7f)) """ % (self.dbType, self.speciesId, self.dbType, self.dbType, self.dbType, mthresh) -# DescriptionText.append(HT.Span(' with a ', HT.U('trans-QTL'), ' having an LRS between %g and %g using a %g Mb exclusion buffer' % (min(lowerLimit, upperLimit), max(lowerLimit, upperLimit), mthresh))) -# query.append(" (%s) " % clauseItem) -# self.orderByDefalut = "LRS" -# else: -# pass -#elif itemCmd.upper() in ("RANGE"): -# #XZ, 03/05/2009: Xiaodong changed Data to ProbeSetData -# clauseItem = " (select Pow(2, max(value) -min(value)) from ProbeSetData where Id = ProbeSetXRef.dataId) > %2.7f and (select Pow(2, max(value) -min(value)) from ProbeSetData where Id = ProbeSetXRef.dataId) < %2.7f " % (min(lowerLimit, upperLimit), max(lowerLimit, upperLimit)) -# query.append(" (%s) " % clauseItem) -# DescriptionText.append(HT.Span(' with a range of expression that varied between %g and %g' % (min(lowerLimit, upperLimit), max(lowerLimit, upperLimit)), " (fold difference)")) -#else: -# clauseItem = " %sXRef.%s > %2.7f and %sXRef.%s < %2.7f " % \ -# (self.dbType, itemCmd, min(lowerLimit, upperLimit), self.dbType, itemCmd, max(lowerLimit, upperLimit)) -# query.append(" (%s) " % clauseItem) -# self.orderByDefalut = itemCmd -# DescriptionText.append(HT.Span(' with ', HT.U(itemCmd), ' between %g and %g' % (min(lowerLimit, upperLimit), max(lowerLimit, upperLimit)))) + return self.real_run(">") class MeanSearch(ProbeSetSearch): @@ -508,7 +454,6 @@ if __name__ == "__main__": import MySQLdb import sys - from base import webqtlConfig from base.data_set import create_dataset from base.templatePage import templatePage @@ -540,11 +485,11 @@ if __name__ == "__main__": ProbeSetXRef.ProbeSetFreezeId = 112""") #print(pf(cursor.fetchall())) - #results = ProbeSetSearch("salt", dataset, cursor, db_conn).run() + results = ProbeSetSearch("shh", None, dataset, cursor, db_conn).run() #results = RifSearch("diabetes", dataset, cursor, db_conn).run() #results = WikiSearch("nicotine", dataset, cursor, db_conn).run() - results = CisLrsSearch(['99'], '>', dataset, cursor, db_conn).run() # cisLRS > 99 - #results = LrsSearch('9', '99', '1', '50', '150', '=', dataset, cursor, db_conn).run() + #results = CisLrsSearch(['99'], '>', dataset, cursor, db_conn).run() # cisLRS > 99 + #results = LrsSearch('99', '>', dataset, cursor, db_conn).run() #results = TransLrsSearch(['9', '999', '10'], dataset, cursor, db_conn).run() #results = PhenotypeSearch("brain", dataset, cursor, db_conn).run() #results = GenotypeSearch("rs13475699", dataset, cursor, db_conn).run() diff --git a/wqflask/wqflask/search_results.py b/wqflask/wqflask/search_results.py index fe091f97..63e0153d 100644 --- a/wqflask/wqflask/search_results.py +++ b/wqflask/wqflask/search_results.py @@ -68,7 +68,7 @@ class SearchResultPage(templatePage): # return ########################################### - # Names and IDs of RISet / F2 set + # Names and IDs of group / F2 set ########################################### # All Phenotypes is a special case we'll deal with later @@ -97,23 +97,23 @@ class SearchResultPage(templatePage): """ self.trait_list = [] + + group = self.dataset.group + species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, group=group) + # result_set represents the results for each search term; a search of # "shh grin2b" would have two sets of results, one for each term print("self.results is:", pf(self.results)) for result in self.results: if not result: continue - - group = self.dataset.group - species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, RISet=group) - + #### Excel file needs to be generated #### print("foo locals are:", locals()) trait_id = result[0] - this_trait = webqtlTrait(self.db_conn, db=self.dataset, name=trait_id) + this_trait = webqtlTrait(self.db_conn, dataset=self.dataset, name=trait_id) this_trait.retrieveInfo(QTL=True) - print("this_trait is:", pf(this_trait)) self.trait_list.append(this_trait) self.dataset.get_trait_info(self.trait_list, species) @@ -134,6 +134,8 @@ class SearchResultPage(templatePage): # We fall back to the dataset type as the key to get the right object search_type = self.dataset.type + print("search_type is:", pf(search_type)) + # This is throwing an error when a_search['key'] is None, so I changed above #search_type = string.upper(a_search['key']) #if not search_type: @@ -146,7 +148,7 @@ class SearchResultPage(templatePage): self.dataset, self.cursor, self.db_conn).run()) - + print("in the search results are:", self.results) diff --git a/wqflask/wqflask/show_trait/show_trait.py b/wqflask/wqflask/show_trait/show_trait.py index 3dac5933..db2636bc 100755 --- a/wqflask/wqflask/show_trait/show_trait.py +++ b/wqflask/wqflask/show_trait/show_trait.py @@ -35,12 +35,12 @@ class ShowTrait(templatePage): self.fd = fd templatePage.__init__(self, fd) - assert self.openMysql(), "No datbase!" + assert self.openMysql(), "No database!" this_trait = self.get_this_trait() ##read genotype file - fd.RISet = this_trait.riset + fd.group = this_trait.group fd.readGenotype() if not fd.genotype: @@ -62,7 +62,7 @@ class ShowTrait(templatePage): # Some fields, like method, are defaulted to None; otherwise in IE the field can't be changed using jquery hddn = OrderedDict( FormID = fmID, - RISet = fd.RISet, + group = fd.group, submitID = '', scale = 'physic', additiveCheck = 'ON', @@ -120,7 +120,7 @@ class ShowTrait(templatePage): hddn['attribute_names'] = "" hddn['mappingMethodId'] = webqtlDatabaseFunction.getMappingMethod (cursor=self.cursor, - groupName=fd.RISet) + groupName=fd.group) if fd.identification: hddn['identification'] = fd.identification @@ -159,8 +159,8 @@ class ShowTrait(templatePage): self.hddn = hddn self.sample_group_types = OrderedDict() - self.sample_group_types['samples_primary'] = fd.RISet + " Only" - self.sample_group_types['samples_other'] = "Non-" + fd.RISet + self.sample_group_types['samples_primary'] = fd.group + " Only" + self.sample_group_types['samples_other'] = "Non-" + fd.group self.sample_group_types['samples_all'] = "All Cases" sample_lists = [group.sample_list for group in self.sample_groups] print("sample_lists is:", pf(sample_lists)) @@ -180,12 +180,12 @@ class ShowTrait(templatePage): trait_id = self.fd['trait_id'] cell_id = self.fd.get('CellID') - this_trait = webqtlTrait(db=dataset, name=trait_id, cellid=cell_id, cursor=self.cursor) + this_trait = webqtlTrait(self.db_conn, db=dataset, name=trait_id, cellid=cell_id) ##identification, etc. self.fd.identification = '%s : %s' % (this_trait.db.shortname, trait_id) this_trait.returnURL = webqtlConfig.CGIDIR + webqtlConfig.SCRIPTFILE + '?FormID=showDatabase&database=%s\ - &ProbeSetID=%s&RISet=%s&parentsf1=on' %(dataset, trait_id, self.fd['RISet']) + &ProbeSetID=%s&group=%s&parentsf1=on' %(dataset, trait_id, self.fd['group']) if cell_id: self.fd.identification = '%s/%s'%(self.fd.identification, cell_id) @@ -198,7 +198,7 @@ class ShowTrait(templatePage): def dispTraitInformation(self, fd, title1Body, hddn, this_trait): - _Species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, RISet=fd.RISet) + _Species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, group=fd.group) #tbl = HT.TableLite(cellpadding=2, Class="collap", style="margin-left:20px;", width="840", valign="top", id="target1") @@ -245,9 +245,9 @@ class ShowTrait(templatePage): else: pass - self.cursor.execute('SELECT Name FROM InbredSet WHERE Name="%s"' % fd.RISet) + self.cursor.execute('SELECT Name FROM InbredSet WHERE Name="%s"' % fd.group) if this_trait: - addSelectionButton = HT.Href(url="#redirect", onClick="addRmvSelection('%s', document.getElementsByName('%s')[0], 'addToSelection');" % (fd.RISet, 'dataInput')) + addSelectionButton = HT.Href(url="#redirect", onClick="addRmvSelection('%s', document.getElementsByName('%s')[0], 'addToSelection');" % (fd.group, 'dataInput')) addSelectionButton_img = HT.Image("/images/add_icon.jpg", name="addselect", alt="Add To Collection", title="Add To Collection", style="border:none;") #addSelectionButton.append(addSelectionButton_img) addSelectionText = "Add" @@ -403,8 +403,8 @@ class ShowTrait(templatePage): probeResult = self.cursor.fetchone() if probeResult[0] > 0: - probeurl = "%s?FormID=showProbeInfo&database=%s&ProbeSetID=%s&CellID=%s&RISet=%s&incparentsf1=ON" \ - % (os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), this_trait.db, this_trait.name, this_trait.cellid, fd.RISet) + probeurl = "%s?FormID=showProbeInfo&database=%s&ProbeSetID=%s&CellID=%s&group=%s&incparentsf1=ON" \ + % (os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), this_trait.db, this_trait.name, this_trait.cellid, fd.group) probeButton = HT.Href(url="#", onClick="javascript:openNewWin('%s'); return false;" % probeurl) probeButton_img = HT.Image("/images/probe_icon.jpg", name="probe", alt=" Check sequence of probes ", title=" Check sequence of probes ", style="border:none;") #probeButton.append(probeButton_img) @@ -430,7 +430,7 @@ class ShowTrait(templatePage): # )) #tSpan = HT.Span(Class="fs13") - #tSpan.append(str(_Species).capitalize(), ", ", fd.RISet) + #tSpan.append(str(_Species).capitalize(), ", ", fd.group) # #tbl.append(HT.TR( # HT.TD('Species and Group: ', Class="fwb fs13", valign="top", nowrap="on"), @@ -805,6 +805,7 @@ class ShowTrait(templatePage): #stats_row = HT.TR() #stats_cell = HT.TD() + # This should still be riset here - Sam - Nov. 2012 if fd.genotype.type == "riset": samplelist = fd.f1list + fd.samplelist else: @@ -839,15 +840,15 @@ class ShowTrait(templatePage): other_samples = map(lambda X:"_2nd_"+X, fd.f1list + fd.parlist) + other_samples #XZ: note that fd.f1list and fd.parlist are added. print("ac1") # This is the one used for first sall3 self.MDP_menu.append(('All Cases','0')) - self.MDP_menu.append(('%s Only' % fd.RISet, '1')) - self.MDP_menu.append(('Non-%s Only' % fd.RISet, '2')) + self.MDP_menu.append(('%s Only' % fd.group, '1')) + self.MDP_menu.append(('Non-%s Only' % fd.group, '2')) else: if (len(other_samples) > 0) and (len(primary_samples) + len(other_samples) > 3): print("ac2") self.MDP_menu.append(('All Cases','0')) - self.MDP_menu.append(('%s Only' % fd.RISet,'1')) - self.MDP_menu.append(('Non-%s Only' % fd.RISet,'2')) + self.MDP_menu.append(('%s Only' % fd.group,'1')) + self.MDP_menu.append(('Non-%s Only' % fd.group,'2')) all_samples = primary_samples all_samples.sort(key=webqtlUtil.natsort_key) all_samples = map(lambda X:"_2nd_"+X, fd.f1list + fd.parlist) + all_samples @@ -895,7 +896,7 @@ class ShowTrait(templatePage): # for sampleNameOrig in all_samples]] # - #Using just the RISet sample + #Using just the group sample for sampleNameOrig in primary_samples: sampleName = sampleNameOrig.replace("_2nd_", "") @@ -908,7 +909,7 @@ class ShowTrait(templatePage): vals2.append(thisValFull) - #Using all non-RISet samples only + #Using all non-group samples only for sampleNameOrig in other_samples: sampleName = sampleNameOrig.replace("_2nd_", "") @@ -951,10 +952,10 @@ class ShowTrait(templatePage): break elif (i == 1 and len(primary_samples) < 4): stats_container = HT.Div(id="stats_tabs%s" % i, Class="ui-tabs") - #stats_container.append(HT.Div(HT.Italic("Fewer than 4 " + fd.RISet + " case data were entered. No statistical analysis has been attempted."))) + #stats_container.append(HT.Div(HT.Italic("Fewer than 4 " + fd.group + " case data were entered. No statistical analysis has been attempted."))) elif (i == 2 and len(other_samples) < 4): stats_container = HT.Div(id="stats_tabs%s" % i, Class="ui-tabs") - stats_container.append(HT.Div(HT.Italic("Fewer than 4 non-" + fd.RISet + " case data were entered. No statistical analysis has been attempted."))) + stats_container.append(HT.Div(HT.Italic("Fewer than 4 non-" + fd.group + " case data were entered. No statistical analysis has been attempted."))) #stats_script_text = """$(function() { $("#stats_tabs0").tabs(); $("#stats_tabs1").tabs(); $("#stats_tabs2").tabs();});""" else: continue @@ -995,7 +996,7 @@ class ShowTrait(templatePage): except: plotTitle = str(this_trait.name) - #normalplot_img = BasicStatisticsFunctions.plotNormalProbability(vals=vals, RISet=fd.RISet, title=plotTitle, specialStrains=specialStrains) + #normalplot_img = BasicStatisticsFunctions.plotNormalProbability(vals=vals, group=fd.group, title=plotTitle, specialStrains=specialStrains) #normalplot.append(HT.TR(HT.TD(normalplot_img))) #normalplot.append(HT.TR(HT.TD(HT.BR(),HT.BR(),"This plot evaluates whether data are \ #normally distributed. Different symbols represent different groups.",HT.BR(),HT.BR(), @@ -1018,7 +1019,7 @@ class ShowTrait(templatePage): #barName_div = HT.Div(id="statstabs-3") #barName_container = HT.Paragraph() #barName = HT.TableLite(cellspacing=0, cellpadding=0, width="100%") - #barName_img = BasicStatisticsFunctions.plotBarGraph(identification=fd.identification, RISet=fd.RISet, vals=vals, type="name") + #barName_img = BasicStatisticsFunctions.plotBarGraph(identification=fd.identification, group=fd.group, vals=vals, type="name") #barName.append(HT.TR(HT.TD(barName_img))) #barName_container.append(barName) #barName_div.append(barName_container) @@ -1027,7 +1028,7 @@ class ShowTrait(templatePage): #barRank_div = HT.Div(id="statstabs-4") #barRank_container = HT.Paragraph() #barRank = HT.TableLite(cellspacing=0, cellpadding=0, width="100%") - #barRank_img = BasicStatisticsFunctions.plotBarGraph(identification=fd.identification, RISet=fd.RISet, vals=vals, type="rank") + #barRank_img = BasicStatisticsFunctions.plotBarGraph(identification=fd.identification, group=fd.group, vals=vals, type="rank") #barRank.append(HT.TR(HT.TD(barRank_img))) #barRank_container.append(barRank) #barRank_div.append(barRank_container) @@ -1048,16 +1049,16 @@ class ShowTrait(templatePage): def build_correlation_tools(self, fd, this_trait): - #species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, RISet=fd.RISet) + #species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, group=fd.group) - RISetgp = fd.RISet + this_group = fd.group # We're checking a string here! - assert isinstance(RISetgp, basestring), "We need a string type thing here" - if RISetgp[:3] == 'BXD': - RISetgp = 'BXD' + assert isinstance(this_group, basestring), "We need a string type thing here" + if this_group[:3] == 'BXD': + this_group = 'BXD' - if RISetgp: + if this_group: #sample_correlation = HT.Input(type='button',name='sample_corr', value=' Compute ', Class="button sample_corr") #lit_correlation = HT.Input(type='button',name='lit_corr', value=' Compute ', Class="button lit_corr") #tissue_correlation = HT.Input(type='button',name='tiss_corr', value=' Compute ', Class="button tiss_corr") @@ -1074,7 +1075,7 @@ class ShowTrait(templatePage): self.cursor.execute('''SELECT PublishFreeze.FullName,PublishFreeze.Name FROM PublishFreeze,InbredSet WHERE PublishFreeze.InbredSetId = InbredSet.Id and InbredSet.Name = %s and PublishFreeze.public > %s''', - (RISetgp, webqtlConfig.PUBLICTHRESH)) + (this_group, webqtlConfig.PUBLICTHRESH)) for item in self.cursor.fetchall(): dataset_menu.append(dict(tissue=None, datasets=[item])) @@ -1082,7 +1083,7 @@ class ShowTrait(templatePage): self.cursor.execute('''SELECT GenoFreeze.FullName,GenoFreeze.Name FROM GenoFreeze, InbredSet WHERE GenoFreeze.InbredSetId = InbredSet.Id and InbredSet.Name = %s and GenoFreeze.public > %s''', - (RISetgp, webqtlConfig.PUBLICTHRESH)) + (this_group, webqtlConfig.PUBLICTHRESH)) for item in self.cursor.fetchall(): dataset_menu.append(dict(tissue=None, datasets=[item])) @@ -1098,7 +1099,7 @@ class ShowTrait(templatePage): InbredSet WHERE ProbeSetFreeze.ProbeFreezeId = ProbeFreeze.Id and ProbeFreeze.TissueId = %s and ProbeSetFreeze.public > %s and ProbeFreeze.InbredSetId = InbredSet.Id and InbredSet.Name like %s order by ProbeSetFreeze.CreateTime desc, ProbeSetFreeze.AvgId ''', - (tissue_id, webqtlConfig.PUBLICTHRESH, "%" + RISetgp + "%")) + (tissue_id, webqtlConfig.PUBLICTHRESH, "%" + this_group + "%")) print("phun8") dataset_sub_menu = [item for item in self.cursor.fetchall() if item] #for item2 in self.cursor.fetchall(): @@ -1257,11 +1258,11 @@ class ShowTrait(templatePage): def dispMappingTools(self, fd, title4Body, this_trait): - _Species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, RISet=fd.RISet) + _Species = webqtlDatabaseFunction.retrieveSpecies(cursor=self.cursor, group=fd.group) - RISetgp = fd.RISet - if RISetgp[:3] == 'BXD': - RISetgp = 'BXD' + this_group = fd.group + if this_group[:3] == 'BXD': + this_group = 'BXD' #check boxes - one for regular interval mapping, the other for composite permCheck1= HT.Input(type='checkbox', Class='checkbox', name='permCheck1',checked="on") @@ -1454,7 +1455,7 @@ class ShowTrait(templatePage): # Treat Interval Mapping and Marker Regression and Pair Scan as a group for displaying #disable Interval Mapping and Marker Regression and Pair Scan for human and the dataset doesn't have genotype file - mappingMethodId = webqtlDatabaseFunction.getMappingMethod(cursor=self.cursor, groupName=RISetgp) + mappingMethodId = webqtlDatabaseFunction.getMappingMethod(cursor=self.cursor, groupName=this_group) mapping_script = HT.Script(language="Javascript") mapping_script_text = """$(function() { $("#mapping_tabs").tabs(); });""" @@ -1526,7 +1527,7 @@ class ShowTrait(templatePage): sample_names=primary_sample_names, this_trait=this_trait, sample_group_type='primary', - header="%s Only" % (fd.RISet)) + header="%s Only" % (fd.group)) other_sample_names = [] for sample in this_trait.data.keys(): @@ -1547,7 +1548,7 @@ class ShowTrait(templatePage): sample_names=other_sample_names, this_trait=this_trait, sample_group_type='other', - header="Non-%s" % (fd.RISet)) + header="Non-%s" % (fd.group)) self.sample_groups = (primary_samples, other_samples) else: diff --git a/wqflask/wqflask/templates/index_page.html b/wqflask/wqflask/templates/index_page.html index a113bc15..c01898b3 100644 --- a/wqflask/wqflask/templates/index_page.html +++ b/wqflask/wqflask/templates/index_page.html @@ -92,8 +92,8 @@ "btn" value="Advanced Search" onclick= "javascript:window.open('/index3.html', '_self');"> - + + diff --git a/wqflask/wqflask/templates/search_result_page.html b/wqflask/wqflask/templates/search_result_page.html index e393ced6..54cdd42b 100644 --- a/wqflask/wqflask/templates/search_result_page.html +++ b/wqflask/wqflask/templates/search_result_page.html @@ -23,7 +23,7 @@ {% if search_terms %}
Mapping Tools
- - - -
-
-
-
+
+
+
+ Marker Regression++ {{ this_trait.name }}: {{ this_trait.description_fmt }} + +
-
@@ -135,8 +136,9 @@
-
diff --git a/wqflask/wqflask/views.py b/wqflask/wqflask/views.py
index 503b0972..f6c0dfb0 100644
--- a/wqflask/wqflask/views.py
+++ b/wqflask/wqflask/views.py
@@ -18,6 +18,7 @@ from flask import render_template, request, make_response, Response, Flask, g, c
from wqflask import search_results
from wqflask.show_trait import show_trait
from wqflask.show_trait import export_trait_data
+from wqflask.marker_regression import marker_regression
from wqflask.correlation import CorrelationPage
from wqflask.dataSharing import SharingInfo, SharingInfoPage
@@ -89,27 +90,6 @@ def whats_new_page():
print("\nnews_item is: %s\n" % (news_item))
return render_template("whats_new.html", news_items=news_items)
-
-@app.route("/show_trait")
-def show_trait_page():
- # Here it's currently too complicated not to use an fd that is a webqtlFormData
- #fd = webqtlFormData.webqtlFormData(request.args)
- #print("stp y1:", pf(vars(fd)))
- template_vars = show_trait.ShowTrait(request.args)
-
- print("js_data before dump:", template_vars.js_data)
-
- template_vars.js_data = json.dumps(template_vars.js_data,
- default=json_default_handler,
- indent=" ")
- # Sorting the keys messes up the ordered dictionary, so don't do that
- #sort_keys=True)
-
- print("js_data after dump:", template_vars.js_data)
-
- print("show_trait template_vars:", pf(template_vars.__dict__))
- return render_template("show_trait.html", **template_vars.__dict__)
-
@app.route('/export_trait_csv', methods=('POST',))
def export_trait_excel():
"""Excel file consisting of the sample data from the trait data and analysis page"""
@@ -150,33 +130,52 @@ def export_trait_csv():
mimetype='text/csv',
headers={"Content-Disposition":"attachment;filename=test.csv"})
+@app.route("/show_trait")
+def show_trait_page():
+ # Here it's currently too complicated not to use an fd that is a webqtlFormData
+ #fd = webqtlFormData.webqtlFormData(request.args)
+ #print("stp y1:", pf(vars(fd)))
+ template_vars = show_trait.ShowTrait(request.args)
+ print("js_data before dump:", template_vars.js_data)
+ template_vars.js_data = json.dumps(template_vars.js_data,
+ default=json_default_handler,
+ indent=" ")
+ # Sorting the keys messes up the ordered dictionary, so don't do that
+ #sort_keys=True)
+
+ print("js_data after dump:", template_vars.js_data)
+ print("show_trait template_vars:", pf(template_vars.__dict__))
+ return render_template("show_trait.html", **template_vars.__dict__)
+
+@app.route("/marker_regression", methods=('POST',))
+def marker_regression_page():
+ template_vars = marker_regression.MarkerRegression(request.form)
+ #print("js_data before dump:", template_vars.js_data)
+ #template_vars.js_data = json.dumps(template_vars.js_data,
+ # default=json_default_handler,
+ # indent=" ")
+ #print("js_data after dump:", template_vars.js_data)
+ print("marker_regression template_vars:", pf(template_vars.__dict__))
+ return render_template("marker_regression.html", **template_vars.__dict__)
@app.route("/corr_compute", methods=('POST',))
def corr_compute_page():
- #print("In corr_compute, request.args is:", pf(request.form))
+ print("In corr_compute, request.args is:", pf(request.form))
fd = webqtlFormData.webqtlFormData(request.form)
- print("Have fd")
template_vars = CorrelationPage.CorrelationPage(fd)
- print("Made it to rendering")
return render_template("correlation_page.html", **template_vars.__dict__)
@app.route("/int_mapping", methods=('POST',))
def interval_mapping_page():
- fd = webqtlFormData.webqtlFormData(request.form)
- print("Have fd")
- template_vars = CorrelationPage.CorrelationPage(fd)
- print("Made it to rendering")
- return render_template("correlation_page.html", **template_vars.__dict__)
-
+ template_vars = interval_mapping.IntervalMapping(request.args)
+ return render_template("interval_mapping.html", **template_vars.__dict__)
# Todo: Can we simplify this? -Sam
def sharing_info_page():
"""Info page displayed when the user clicks the "Info" button next to the dataset selection"""
print("In sharing_info_page")
fd = webqtlFormData.webqtlFormData(request.args)
- print("2Have fd")
template_vars = SharingInfoPage.SharingInfoPage(fd)
- print("2 Made it to rendering")
return template_vars
--
cgit v1.2.3
From 91ed29ef68e8ad29b728f7f574ccc83730d9f7ab Mon Sep 17 00:00:00 2001
From: Zachary Sloan
Date: Thu, 3 Jan 2013 18:15:32 -0600
Subject: Began working on marker_regression.py and created Chromosomes class
in species.py
---
wqflask/base/data_set.py | 6 +-
wqflask/base/species.py | 52 +-
wqflask/wqflask/interval_analyst/GeneUtil.py | 124 ++
.../interval_analyst/IntervalAnalystPage.py | 405 +++++
wqflask/wqflask/interval_analyst/__init__.py | 0
.../marker_regression/MarkerRegressionPage.py | 1648 ++++++++++++++++++++
wqflask/wqflask/marker_regression/__init__.py | 0
.../wqflask/marker_regression/marker_regression.py | 1648 ++++++++++++++++++++
8 files changed, 3870 insertions(+), 13 deletions(-)
create mode 100755 wqflask/wqflask/interval_analyst/GeneUtil.py
create mode 100755 wqflask/wqflask/interval_analyst/IntervalAnalystPage.py
create mode 100644 wqflask/wqflask/interval_analyst/__init__.py
create mode 100644 wqflask/wqflask/marker_regression/MarkerRegressionPage.py
create mode 100644 wqflask/wqflask/marker_regression/__init__.py
create mode 100755 wqflask/wqflask/marker_regression/marker_regression.py
(limited to 'wqflask/base/data_set.py')
diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py
index 36d4acaf..50ef8f57 100755
--- a/wqflask/base/data_set.py
+++ b/wqflask/base/data_set.py
@@ -85,8 +85,8 @@ class DatasetGroup(object):
self.f1list = None
self.parlist = None
self.allsamples = None
-
-
+
+
#def read_genotype(self):
# self.read_genotype_file()
#
@@ -158,8 +158,8 @@ class DataSet(object):
self.retrieve_other_names()
- self.species = species.TheSpecies(self)
self.group = DatasetGroup(self) # sets self.group and self.group_id and gets genotype
+ self.species = species.TheSpecies(self)
diff --git a/wqflask/base/species.py b/wqflask/base/species.py
index 98941ce5..1fd76772 100644
--- a/wqflask/base/species.py
+++ b/wqflask/base/species.py
@@ -1,16 +1,48 @@
-from __future__ import print_function, division
+from __future__ import absolute_import, print_function, division
+import collections
+
+from flask import Flask, g
+
+#from MySQLdb import escape_string as escape
+
+from pprint import pformat as pf
class TheSpecies(object):
def __init__(self, dataset):
self.dataset = dataset
+ print("self.dataset is:", pf(self.dataset.__dict__))
+ self.chromosomes = Chromosomes(self.dataset.group.name)
+
+ #@property
+ #def chromosomes(self):
+ # chromosomes = [("All", -1)]
+ #
+ # for counter, genotype in enumerate(self.dataset.group.genotype):
+ # if len(genotype) > 1:
+ # chromosomes.append((genotype.name, counter))
+ #
+ # print("chromosomes is: ", pf(chromosomes))
+ #
+ # return chromosomes
+
+
+
+class Chromosomes(object):
+ def __init__(self, group_name):
+ self.chromosomes = collections.OrderedDict()
+
+ results = g.db.execute("""
+ Select
+ Chr_Length.Name, Length from Chr_Length, InbredSet
+ where
+ Chr_Length.SpeciesId = InbredSet.SpeciesId AND
+ InbredSet.Name = %s
+ Order by OrderId
+ """, group_name).fetchall()
+ print("bike:", results)
+
+ for item in results:
+ self.chromosomes[item.Name] = item.Length
- @property
- def chromosomes(self):
- chromosomes = [("All", -1)]
-
- for counter, genotype in enumerate(self.dataset.group.genotype):
- if len(genotype) > 1:
- chromosomes.append((genotype.name, counter))
-
- return chromosomes
+ print("self.chromosomes:", self.chromosomes)
diff --git a/wqflask/wqflask/interval_analyst/GeneUtil.py b/wqflask/wqflask/interval_analyst/GeneUtil.py
new file mode 100755
index 00000000..43008ecf
--- /dev/null
+++ b/wqflask/wqflask/interval_analyst/GeneUtil.py
@@ -0,0 +1,124 @@
+# Copyright (C) University of Tennessee Health Science Center, Memphis, TN.
+#
+# This program is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Affero General Public License
+# as published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# See the GNU Affero General Public License for more details.
+#
+# This program is available from Source Forge: at GeneNetwork Project
+# (sourceforge.net/projects/genenetwork/).
+#
+# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010)
+# at rwilliams@uthsc.edu and xzhou15@uthsc.edu
+#
+#
+#
+# This module is used by GeneNetwork project (www.genenetwork.org)
+#
+# Created by GeneNetwork Core Team 2010/08/10
+#
+# Last updated by GeneNetwork Core Team 2010/10/20
+
+import string
+
+#Just return a list of dictionaries
+#each dictionary contains sub-dictionary
+def loadGenes(cursor, chrName, diffCol, startMb, endMb, webqtlDb =None, species='mouse'):
+ #cursor.execute("desc GeneList")
+ #results = cursor.fetchall()
+ #fetchFields = map(lambda X:X[0], results)
+ fetchFields = ['SpeciesId', 'Id', 'GeneSymbol', 'GeneDescription', 'Chromosome', 'TxStart', 'TxEnd',
+ 'Strand', 'GeneID', 'NM_ID', 'kgID', 'GenBankID', 'UnigenID', 'ProteinID', 'AlignID',
+ 'exonCount', 'exonStarts', 'exonEnds', 'cdsStart', 'cdsEnd']
+
+ ##List All Species in the Gene Table
+ speciesDict = {}
+ cursor.execute("select Species.Name, GeneList.SpeciesId from Species, GeneList where \
+ GeneList.SpeciesId = Species.Id group by GeneList.SpeciesId")
+ results = cursor.fetchall()
+ for item in results:
+ speciesDict[item[0]] = item[1]
+
+ ##List current Species and other Species
+ speciesId = speciesDict[species]
+ otherSpecies = map(lambda X: [X, speciesDict[X]], speciesDict.keys())
+ otherSpecies.remove([species, speciesId])
+
+ cursor.execute("""SELECT %s from GeneList
+ where
+ SpeciesId = %d AND Chromosome = '%s' AND
+ ((TxStart > %f and TxStart <= %f) OR (TxEnd > %f and TxEnd <= %f))
+ order by txStart
+ """
+ % (string.join(fetchFields, ", "), speciesId, chrName, startMb, endMb, startMb, endMb))
+ results = cursor.fetchall()
+ GeneList = []
+
+ if results:
+ for result in results:
+ newdict = {}
+ for j, item in enumerate(fetchFields):
+ newdict[item] = result[j]
+ #count SNPs if possible
+ if diffCol and species=='mouse':
+ cursor.execute("""
+ select
+ count(*) from BXDSnpPosition
+ where
+ Chr = '%s' AND Mb >= %2.6f AND Mb < %2.6f AND
+ StrainId1 = %d AND StrainId2 = %d
+ """ % (chrName, newdict["TxStart"], newdict["TxEnd"], diffCol[0], diffCol[1]))
+ newdict["snpCount"] = cursor.fetchone()[0]
+ newdict["snpDensity"] = newdict["snpCount"]/(newdict["TxEnd"]-newdict["TxStart"])/1000.0
+ else:
+ newdict["snpDensity"] = newdict["snpCount"] = 0
+
+ try:
+ newdict['GeneLength'] = 1000.0*(newdict['TxEnd'] - newdict['TxStart'])
+ except:
+ pass
+
+ #load gene from other Species by the same name
+ for item in otherSpecies:
+ othSpec, othSpecId = item
+ newdict2 = {}
+
+ cursor.execute("SELECT %s from GeneList where SpeciesId = %d and geneSymbol= '%s' limit 1" %
+ (string.join(fetchFields, ", "), othSpecId, newdict["GeneSymbol"]))
+ resultsOther = cursor.fetchone()
+ if resultsOther:
+ for j, item in enumerate(fetchFields):
+ newdict2[item] = resultsOther[j]
+
+ #count SNPs if possible, could be a separate function
+ if diffCol and othSpec == 'mouse':
+ cursor.execute("""
+ select
+ count(*) from BXDSnpPosition
+ where
+ Chr = '%s' AND Mb >= %2.6f AND Mb < %2.6f AND
+ StrainId1 = %d AND StrainId2 = %d
+ """ % (chrName, newdict["TxStart"], newdict["TxEnd"], diffCol[0], diffCol[1]))
+
+ newdict2["snpCount"] = cursor.fetchone()[0]
+ newdict2["snpDensity"] = newdict2["snpCount"]/(newdict2["TxEnd"]-newdict2["TxStart"])/1000.0
+ else:
+ newdict2["snpDensity"] = newdict2["snpCount"] = 0
+
+ try:
+ newdict2['GeneLength'] = 1000.0*(newdict2['TxEnd'] - newdict2['TxStart'])
+ except:
+ pass
+
+ newdict['%sGene' % othSpec] = newdict2
+
+ GeneList.append(newdict)
+
+ return GeneList
+
+
diff --git a/wqflask/wqflask/interval_analyst/IntervalAnalystPage.py b/wqflask/wqflask/interval_analyst/IntervalAnalystPage.py
new file mode 100755
index 00000000..ec9aa29c
--- /dev/null
+++ b/wqflask/wqflask/interval_analyst/IntervalAnalystPage.py
@@ -0,0 +1,405 @@
+# Copyright (C) University of Tennessee Health Science Center, Memphis, TN.
+#
+# This program is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Affero General Public License
+# as published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# See the GNU Affero General Public License for more details.
+#
+# This program is available from Source Forge: at GeneNetwork Project
+# (sourceforge.net/projects/genenetwork/).
+#
+# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010)
+# at rwilliams@uthsc.edu and xzhou15@uthsc.edu
+#
+#
+#
+# This module is used by GeneNetwork project (www.genenetwork.org)
+#
+# Created by GeneNetwork Core Team 2010/08/10
+#
+# Last updated by GeneNetwork Core Team 2010/10/20
+
+from mod_python import apache, util, Cookie
+import os
+import time
+import pyXLWriter as xl
+
+from htmlgen import HTMLgen2 as HT
+
+import GeneUtil
+from base.templatePage import templatePage
+from utility import webqtlUtil
+from base import webqtlConfig
+
+
+class IntervalAnalystPage(templatePage):
+ filename = webqtlUtil.genRandStr("Itan_")
+
+ _scriptfile = "main.py?FormID=intervalAnalyst"
+
+ #A dictionary that lets us map the html form names "txStart_mm6" -> "Mb Start (mm8)"
+ #the first item is the short name (column headers) and the second item is the long name (dropdown list)
+ # [short name, long name, category]
+ columnNames = {"GeneSymbol" : ["Gene", "Gene Name", 'gene'],
+ "GeneDescription" : ["Description", "Gene Description", 'species'],
+ 'GeneNeighborsCount' : ["Neighbors", "Gene Neighbors", 'gene'],
+ 'GeneNeighborsRange' : ["Neighborhood", "Gene Neighborhood (Mb)", 'gene'],
+ 'GeneNeighborsDensity' : ["Gene Density", "Gene Density (Neighbors/Mb)", 'gene'],
+ "ProteinID" : ["Prot ID", "Protein ID", 'protein'],
+ "Chromosome" : ["Chr", "Chromosome", 'species'],
+ "TxStart" : ["Start", "Mb Start", 'species'],
+ "TxEnd" : ["End", "Mb End", 'species'],
+ "GeneLength" : ["Length", "Kb Length", 'species'],
+ "cdsStart" : ["CDS Start", "Mb CDS Start", 'species'],
+ "cdsEnd" : ["CDS End", "Mb CDS End", 'species'],
+ "exonCount" : ["Num Exons", "Exon Count", 'species'],
+ "exonStarts" : ["Exon Starts", "Exon Starts", 'species'],
+ "exonEnds" : ["Exon Ends", "Exon Ends", 'species'],
+ "Strand" : ["Strand", "Strand", 'species'],
+ "GeneID" : ["Gene ID", "Gene ID", 'species'],
+ "GenBankID" : ["GenBank", "GenBank ID", 'species'],
+ "UnigenID" : ["Unigen", "Unigen ID", 'species'],
+ "NM_ID" : ["NM ID", "NM ID", 'species'],
+ "kgID" : ["kg ID", "kg ID", 'species'],
+ "snpCount" : ["SNPs", "SNP Count", 'species'],
+ "snpDensity" : ["SNP Density", "SNP Density", 'species'],
+ "lrs" : ["LRS", "Likelihood Ratio Statistic", 'misc'],
+ "lod" : ["LOD", "Likelihood Odds Ratio", 'misc'],
+ "pearson" : ["Pearson", "Pearson Product Moment", 'misc'],
+ "literature" : ["Lit Corr", "Literature Correlation", 'misc'],
+ }
+
+ ###Species Freeze
+ speciesFreeze = {'mouse':'mm9', 'rat':'rn3', 'human':'hg19'}
+ for key in speciesFreeze.keys():
+ speciesFreeze[speciesFreeze[key]] = key
+
+ def __init__(self, fd):
+
+ templatePage.__init__(self, fd)
+
+ fd.formdata['remote_ip'] = fd.remote_ip
+ if not self.openMysql():
+ return
+
+ self.species = fd.formdata.getvalue("species", "mouse")
+ try:
+ self.startMb = float(fd.formdata.getvalue("startMb"))
+ except:
+ self.startMb = 10
+ try:
+ self.endMb = float(fd.formdata.getvalue("endMb"))
+ except:
+ self.endMb = self.startMb + 10
+
+ self.Chr = fd.formdata.getvalue("chromosome", "1")
+ self.xls = fd.formdata.getvalue("xls", "1")
+ try:
+ s1 = int(fd.formdata.getvalue("s1"))
+ s2 = int(fd.formdata.getvalue("s2"))
+ self.diffColDefault = self.diffCol = [s1, s2]
+ except:
+ self.diffColDefault = self.diffCol = []
+ if self.species != 'mouse':
+ self.diffColDefault = [2, 3]#default is B6 and D2 for other species
+
+ controlFrm, dispFields = self.genControlForm(fd)
+ geneTable, filename = self.genGeneTable(fd, dispFields)
+
+ infoTD = HT.TD(width=400, valign= "top")
+ infoTD.append(HT.Paragraph("Interval Analyst : Chr %s" % self.Chr, Class="title"),
+ HT.Strong("Species : "), self.species.title(), HT.BR(),
+ HT.Strong("Database : "), "UCSC %s" % self.speciesFreeze[self.species], HT.BR(),
+ HT.Strong("Range : "), "%2.6f Mb - %2.6f Mb" % (self.startMb, self.endMb), HT.BR(),
+ )
+ if filename:
+ infoTD.append(HT.BR(), HT.BR(), HT.Href(text="Download", url = "/tmp/" + filename, Class="normalsize")
+ , " output in MS excel format.")
+
+ mainTable = HT.TableLite(HT.TR(infoTD, HT.TD(controlFrm, Class="doubleBorder", width=400), HT.TD(" ", width="")), cellpadding=10)
+ mainTable.append(HT.TR(HT.TD(geneTable, colspan=3)))
+ self.dict['body'] = HT.TD(mainTable)
+ self.dict['title'] = "Interval Analyst"
+
+ def genGeneTable(self, fd, dispFields):
+ filename = ""
+ if self.xls:
+ #import pyXLWriter as xl
+ filename = "IntAn_Chr%s_%2.6f-%2.6f" % (self.Chr, self.startMb, self.endMb)
+ filename += ".xls"
+
+ # Create a new Excel workbook
+ workbook = xl.Writer(os.path.join(webqtlConfig.TMPDIR, filename))
+ worksheet = workbook.add_worksheet()
+ titleStyle = workbook.add_format(align = 'left', bold = 0, size=18, border = 1, border_color="gray")
+ headingStyle = workbook.add_format(align = 'center', bold = 1, size=13, fg_color = 0x1E, color="white", border = 1, border_color="gray")
+
+ ##Write title Info
+ worksheet.write([0, 0], "GeneNetwork Interval Analyst Table", titleStyle)
+ worksheet.write([1, 0], "%s%s" % (webqtlConfig.PORTADDR, os.path.join(webqtlConfig.CGIDIR, self._scriptfile)))
+ #
+ worksheet.write([2, 0], "Date : %s" % time.strftime("%B %d, %Y", time.gmtime()))
+ worksheet.write([3, 0], "Time : %s GMT" % time.strftime("%H:%M ", time.gmtime()))
+ worksheet.write([4, 0], "Search by : %s" % fd.formdata['remote_ip'])
+ worksheet.write([5, 0], "view region : Chr %s %2.6f - %2.6f Mb" % (self.Chr, self.startMb, self.endMb))
+ nTitleRow = 7
+
+ geneTable = HT.TableLite(Class="collap", cellpadding=5)
+ headerRow = HT.TR(HT.TD(" ", Class="fs13 fwb ffl b1 cw cbrb", width="1"))
+ if self.xls:
+ worksheet.write([nTitleRow, 0], "Index", headingStyle)
+
+ for ncol, column in enumerate(dispFields):
+ if len(column) == 1:
+ headerRow.append(HT.TD(self.columnNames[column[0]][0], Class="fs13 fwb ffl b1 cw cbrb", NOWRAP=1,align="Center"))
+ if self.xls:
+ colTitle = self.columnNames[column[0]][0]
+ worksheet.write([nTitleRow, ncol+1], colTitle, headingStyle)
+ worksheet.set_column([ncol+1, ncol+1], 2*len(colTitle))
+ else:
+ headerRow.append(HT.TD(self.columnNames[column[0]][0], HT.BR(), " (%s)" % self.speciesFreeze[column[1]],
+ Class="fs13 fwb ffl b1 cw cbrb", NOWRAP=1, align="Center"))
+ if self.xls:
+ colTitle = self.columnNames[column[0]][0] + " (%s)" % self.speciesFreeze[column[1]]
+ worksheet.write([nTitleRow, ncol+1], colTitle, headingStyle)
+ worksheet.set_column([ncol+1, ncol+1], 2*len(colTitle))
+ #headerRow.append(HT.TD(self.columnNames[column[0]][0], HT.BR(),
+ # "(%s %s)" % (column[1].title(), self.speciesFreeze[column[1]]),
+ # Class="colorBlue", NOWRAP=1, align="Center"))
+ geneTable.append(headerRow)
+
+ geneCol = GeneUtil.loadGenes(self.cursor, self.Chr, self.diffColDefault, self.startMb, self.endMb, species=self.species)
+ for gIndex, theGO in enumerate(geneCol):
+ geneRow = HT.TR(HT.TD(gIndex+1, Class="fs12 fwn b1", align="right"))
+ if self.xls:
+ nTitleRow += 1
+ worksheet.write([nTitleRow, 0], gIndex + 1)
+
+ for ncol, column in enumerate(dispFields):
+ if len(column) == 1 or column[1]== self.species:
+ keyValue = ""
+ fieldName = column[0]
+ curSpecies = self.species
+ curGO = theGO
+ if theGO.has_key(fieldName):
+ keyValue = theGO[fieldName]
+ else:
+ fieldName , othSpec = column
+ curSpecies = othSpec
+ subGO = '%sGene' % othSpec
+ keyValue = ""
+ curGO = theGO[subGO]
+ if theGO[subGO].has_key(fieldName):
+ keyValue = theGO[subGO][fieldName]
+
+ if self.xls:
+ worksheet.write([nTitleRow, ncol+1], keyValue)
+ geneRow.append(self.formatTD(keyValue, fieldName, curSpecies, curGO))
+
+ geneTable.append(geneRow)
+
+ if self.xls:
+ workbook.close()
+ return geneTable, filename
+
+ def formatTD(self, keyValue, fieldName, Species, theGO):
+ if keyValue is None:
+ keyValue = ""
+ if keyValue != "":
+ if fieldName in ("exonStarts", "exonEnds"):
+ keyValue = string.replace(keyValue, ',', ' ')
+ return HT.TD(HT.Span(keyValue, Class="code", Id="green"), width=350, Class="fs12 fwn b1")
+ elif fieldName in ("GeneDescription"):
+ if keyValue == "---":
+ keyValue = ""
+ return HT.TD(keyValue, Class="fs12 fwn b1", width=300)
+ elif fieldName in ("GeneSymbol"):
+ webqtlLink = HT.Href("./%s?cmd=sch&gene=%s&alias=1&species=%s" % (webqtlConfig.SCRIPTFILE, keyValue, Species),
+ HT.Image("/images/webqtl_search.gif", border=0, valign="top"), target="_blank")
+ if theGO['GeneID']:
+ geneSymbolLink = HT.Href(webqtlConfig.NCBI_LOCUSID % theGO['GeneID'], keyValue, Class="normalsize", target="_blank")
+ else:
+ geneSymbolLink = keyValue
+ return HT.TD(webqtlLink, geneSymbolLink, Class="fs12 fwn b1",NOWRAP=1)
+ elif fieldName == 'UnigenID':
+ try:
+ gurl = HT.Href(webqtlConfig.UNIGEN_ID % tuple(string.split(keyValue,'.')[:2]), keyValue, Class="normalsize", target="_blank")
+ except:
+ gurl = keyValue
+ return HT.TD(gurl, Class="fs12 fwn b1",NOWRAP=1)
+ elif fieldName in ("exonCount", "Chromosome"):
+ return HT.TD(keyValue, Class="fs12 fwn b1",align="right")
+ elif fieldName in ("snpCount"):
+ if keyValue:
+ snpString = HT.Href(url="%s&chr=%s&start=%s&end=%s&geneName=%s&s1=%d&s2=%d" % (os.path.join(webqtlConfig.CGIDIR, 'main.py?FormID=snpBrowser'),
+ theGO["Chromosome"], theGO["TxStart"], theGO["TxEnd"], theGO["GeneSymbol"], self.diffColDefault[0], self.diffColDefault[1]),
+ text=theGO["snpCount"], target="_blank", Class="normalsize")
+ else:
+ snpString = keyValue
+ return HT.TD(snpString, Class="fs12 fwn b1",align="right")
+ elif fieldName in ("snpDensity", "GeneLength"):
+ if keyValue: keyValue = "%2.3f" % keyValue
+ else: keyValue = ""
+ return HT.TD(keyValue, Class="fs12 fwn b1",align="right")
+ elif fieldName in ("TxStart", "TxEnd"):
+ return HT.TD("%2.6f" % keyValue, Class="fs12 fwn b1",align="right")
+ else:
+ return HT.TD(keyValue, Class="fs12 fwn b1",NOWRAP=1)
+ else:
+ return HT.TD(keyValue, Class="fs12 fwn b1",NOWRAP=1,align="right")
+
+ def genControlForm(self, fd):
+ ##desc GeneList
+ self.cursor.execute("Desc GeneList")
+ GeneListFields = self.cursor.fetchall()
+ GeneListFields = map(lambda X: X[0], GeneListFields)
+
+ #group columns by category--used for creating the dropdown list of possible columns
+ categories = {}
+ for item in self.columnNames.keys():
+ category = self.columnNames[item]
+ if category[-1] not in categories.keys():
+ categories[category[-1]] = [item ]
+ else:
+ categories[category[-1]] = categories[category[-1]]+[item]
+
+ ##List All Species in the Gene Table
+ speciesDict = {}
+ self.cursor.execute("select Species.Name, GeneList.SpeciesId from Species, GeneList where \
+ GeneList.SpeciesId = Species.Id group by GeneList.SpeciesId order by Species.Id")
+ results = self.cursor.fetchall()
+ speciesField = categories.pop('species', [])
+ categoriesOrder = ['gene', 'protein']
+ for item in results:
+ specName, specId = item
+ categoriesOrder.append(specName)
+ speciesDict[specName] = specId
+ AppliedField = []
+ for item2 in speciesField:
+ if item2 in GeneListFields:
+ self.cursor.execute("select %s from GeneList where SpeciesId = %d and %s is not NULL limit 1 " % (item2, specId, item2))
+ columnApply = self.cursor.fetchone()
+ if not columnApply:
+ continue
+ elif specName != 'mouse' and item2 in ('snpCount', 'snpDensity'):
+ continue
+ else:
+ pass
+ AppliedField.append(item2)
+ categories[specName] = AppliedField
+
+ categoriesOrder += ['misc']
+
+ ############################################################
+ ## Create the list of possible columns for the dropdown list
+ ############################################################
+ allColumnsList = HT.Select(name="allColumns", Class="snpBrowserDropBox")
+
+ for category in categoriesOrder:
+ allFields = categories[category]
+ if allFields:
+ geneOpt = HT.Optgroup(label=category.title())
+ for item in allFields:
+ if category in self.speciesFreeze.keys():
+ geneOpt.append(("%s (%s %s)" % (self.columnNames[item][1], category.title(), self.speciesFreeze[category]),
+ "%s__%s" % (item, self.speciesFreeze[category])))
+ else:
+ geneOpt.append((self.columnNames[item][1], item))
+ geneOpt.sort()
+ allColumnsList.append(geneOpt)
+
+ ######################################
+ ## Create the list of selected columns
+ ######################################
+
+ #cols contains the value of all the selected columns
+ submitCols = cols = fd.formdata.getvalue("columns", "default")
+
+ if cols == "default":
+ if self.species=="mouse": #these are the same columns that are shown on intervalPage.py
+ cols = ['GeneSymbol', 'GeneDescription', 'Chromosome', 'TxStart', 'Strand', 'GeneLength', 'GeneID', 'NM_ID', 'snpCount', 'snpDensity']
+ elif self.species=="rat":
+ cols = ['GeneSymbol', 'GeneDescription', 'Chromosome', 'TxStart', 'GeneLength', 'Strand', 'GeneID', 'UnigenID']
+ else:
+ #should not happen
+ cols = []
+ else:
+ if type(cols)==type(""):
+ cols = [cols]
+
+ colsLst = []
+ dispFields = []
+ for column in cols:
+ if submitCols == "default" and column not in ('GeneSymbol') and (column in GeneListFields or column in speciesField):
+ colsLst.append(("%s (%s %s)" % (self.columnNames[column][1], self.species.title(), self.speciesFreeze[self.species]),
+ "%s__%s" % (column, self.speciesFreeze[self.species])))
+ dispFields.append([column, self.species])
+ else:
+ column2 = column.split("__")
+ if len(column2) == 1:
+ colsLst.append((self.columnNames[column2[0]][1], column))
+ dispFields.append([column])
+ else:
+ thisSpecies = self.speciesFreeze[column2[1]]
+ colsLst.append(("%s (%s %s)" % (self.columnNames[column2[0]][1], thisSpecies.title(), column2[1]),
+ column))
+ dispFields.append((column2[0], thisSpecies))
+ selectedColumnsList = HT.Select(name="columns", Class="snpBrowserSelectBox", multiple="true", data=colsLst, size=6)
+
+ ##########################
+ ## Create the columns form
+ ##########################
+ columnsForm = HT.Form(name="columnsForm", submit=HT.Input(type='hidden'), cgi=os.path.join(webqtlConfig.CGIDIR, self._scriptfile), enctype="multipart/form-data")
+ columnsForm.append(HT.Input(type="hidden", name="fromdatabase", value= fd.formdata.getvalue("fromdatabase", "unknown")))
+ columnsForm.append(HT.Input(type="hidden", name="species", value=self.species))
+ if self.diffCol:
+ columnsForm.append(HT.Input(type="hidden", name="s1", value=self.diffCol[0]))
+ columnsForm.append(HT.Input(type="hidden", name="s2", value=self.diffCol[1]))
+ startBox = HT.Input(type="text", name="startMb", value=self.startMb, size=10)
+ endBox = HT.Input(type="text", name="endMb", value=self.endMb, size=10)
+ addButton = HT.Input(type="button", name="add", value="Add", Class="button", onClick="addToList(this.form.allColumns.options[this.form.allColumns.selectedIndex].text, this.form.allColumns.options[this.form.allColumns.selectedIndex].value, this.form.columns)")
+ removeButton = HT.Input(type="button", name="remove", value="Remove", Class="button", onClick="removeFromList(this.form.columns.selectedIndex, this.form.columns)")
+ upButton = HT.Input(type="button", name="up", value="Up", Class="button", onClick="swapOptions(this.form.columns.selectedIndex, this.form.columns.selectedIndex-1, this.form.columns)")
+ downButton = HT.Input(type="button", name="down", value="Down", Class="button", onClick="swapOptions(this.form.columns.selectedIndex, this.form.columns.selectedIndex+1, this.form.columns)")
+ clearButton = HT.Input(type="button", name="clear", value="Clear", Class="button", onClick="deleteAllElements(this.form.columns)")
+ submitButton = HT.Input(type="submit", value="Refresh", Class="button", onClick="selectAllElements(this.form.columns)")
+
+ selectChrBox = HT.Select(name="chromosome")
+ self.cursor.execute("""
+ Select
+ Chr_Length.Name, Length from Chr_Length, Species
+ where
+ Chr_Length.SpeciesId = Species.Id AND
+ Species.Name = '%s'
+ Order by
+ Chr_Length.OrderId
+ """ % self.species)
+
+ results = self.cursor.fetchall()
+ for chrInfo in results:
+ selectChrBox.append((chrInfo[0], chrInfo[0]))
+ selectChrBox.selected.append(self.Chr)
+
+ innerColumnsTable = HT.TableLite(border=0, Class="collap", cellpadding = 2)
+ innerColumnsTable.append(HT.TR(HT.TD(selectedColumnsList)),
+ HT.TR(HT.TD(clearButton, removeButton, upButton, downButton)))
+ columnsTable = HT.TableLite(border=0, cellpadding=2, cellspacing=0)
+ columnsTable.append(HT.TR(HT.TD(HT.Font("Chr: ", size=-1)),
+ HT.TD(selectChrBox, submitButton)),
+ HT.TR(HT.TD(HT.Font("View: ", size=-1)),
+ HT.TD(startBox, HT.Font("Mb to ", size=-1), endBox, HT.Font("Mb", size=-1))),
+ HT.TR(HT.TD(HT.Font("Show: ", size=-1)),
+ HT.TD(allColumnsList, addButton)),
+ HT.TR(HT.TD(""),
+ HT.TD(innerColumnsTable)))
+ columnsForm.append(columnsTable)
+ #columnsForm.append(HT.Input(type="hidden", name="sort", value=diffCol),
+ # HT.Input(type="hidden", name="identification", value=identification),
+ # HT.Input(type="hidden", name="traitInfo", value=traitInfo))
+
+ return columnsForm, dispFields
diff --git a/wqflask/wqflask/interval_analyst/__init__.py b/wqflask/wqflask/interval_analyst/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/wqflask/wqflask/marker_regression/MarkerRegressionPage.py b/wqflask/wqflask/marker_regression/MarkerRegressionPage.py
new file mode 100644
index 00000000..d02d80b3
--- /dev/null
+++ b/wqflask/wqflask/marker_regression/MarkerRegressionPage.py
@@ -0,0 +1,1648 @@
+# Copyright (C) University of Tennessee Health Science Center, Memphis, TN.
+#
+# This program is free software: you can redistribute it and/or modify it
+# under the terms of the GNU Affero General Public License
+# as published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# See the GNU Affero General Public License for more details.
+#
+# This program is available from Source Forge: at GeneNetwork Project
+# (sourceforge.net/projects/genenetwork/).
+#
+# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010)
+# at rwilliams@uthsc.edu and xzhou15@uthsc.edu
+#
+# This module is used by GeneNetwork project (www.genenetwork.org)
+#
+# Created by GeneNetwork Core Team 2010/08/10
+#
+# Last updated by GeneNetwork Core Team 2010/10/20
+
+import time
+import string
+import math
+from math import *
+import piddle as pid
+import sys,os
+import httplib, urllib
+
+from htmlgen import HTMLgen2 as HT
+from utility import Plot
+from intervalAnalyst import GeneUtil
+from base.webqtlTrait import webqtlTrait
+from base.templatePage import templatePage
+from utility import webqtlUtil
+from base import webqtlConfig
+from dbFunction import webqtlDatabaseFunction
+from base.GeneralObject import GeneralObject
+
+import reaper
+import cPickle
+from utility.THCell import THCell
+from utility.TDCell import TDCell
+
+class MarkerRegressionPage(templatePage):
+
+ def __init__(self, fd):
+
+ templatePage.__init__(self, fd)
+
+ if not self.openMysql():
+ return
+
+ self.initializeParameters(fd)
+
+ filename= webqtlUtil.genRandStr("Itvl_")
+ ChrList,ChrNameOrderIdDict,ChrOrderIdNameDict,ChrLengthMbList= self.getChrNameOrderIdLength(RISet=fd.RISet)
+
+ if self.mappingMethodId == '4': # For PLINK
+
+ traitInfoList = string.split(string.strip(fd.identification),':')
+ probesetName = string.strip(traitInfoList[-1])
+ plinkOutputFileName= webqtlUtil.genRandStr("%s_%s_"%(fd.RISet,probesetName))
+
+ # get related values from fd.allTraitData; the format of 'allTraitValueDict'is {strainName1: value=-0.2...}
+ fd.readData()
+ allTraitValueDict = fd.allTraitData
+
+ #automatically generate pheno txt file for PLINK
+ self.genPhenoTxtFileForPlink(phenoFileName=plinkOutputFileName,RISetName=fd.RISet,probesetName=probesetName, valueDict=allTraitValueDict)
+ # os.system full path is required for input and output files; specify missing value is -9999
+ plink_command = '%splink/plink --noweb --ped %splink/%s.ped --no-fid --no-parents --no-sex --no-pheno --map %splink/%s.map --pheno %s/%s.txt --pheno-name %s --missing-phenotype -9999 --out %s%s --assoc ' % (webqtlConfig.HTMLPATH, webqtlConfig.HTMLPATH, fd.RISet, webqtlConfig.HTMLPATH, fd.RISet, webqtlConfig.TMPDIR, plinkOutputFileName, probesetName, webqtlConfig.TMPDIR, plinkOutputFileName)
+
+ os.system(plink_command)
+
+ if fd.identification:
+ heading2 = HT.Paragraph('Trait ID: %s' % fd.identification)
+ heading2.__setattr__("class","subtitle")
+ self.dict['title'] = '%s: Genome Association' % fd.identification
+ else:
+ heading2 = ""
+ self.dict['title'] = 'Genome Association'
+
+ if fd.traitInfo:
+ symbol,chromosome,MB = string.split(fd.traitInfo,'\t')
+ heading3 = HT.Paragraph('[ ',HT.Strong(HT.Italic('%s' % symbol,id="green")),' on Chr %s @ %s Mb ]' % (chromosome,MB))
+ else:
+ heading3 = ""
+
+ heading = HT.Paragraph('Trait Data Entered for %s Set' % fd.RISet)
+ heading.__setattr__("class","title")
+
+ # header info part:Trait Data Entered for HLC Set & Trait ID:
+ headerdiv = HT.TR(HT.TD(heading, heading2,heading3, width='45%',valign='top', align='left', bgColor='#eeeeee'))
+
+ self.ChrList=ChrList # get chr name from '1' to 'X'
+ self.ChrLengthMbList = ChrLengthMbList
+
+ # build plink result dict based on chr, key is chr name, value is in list type including Snpname, bp and pvalue info
+ plinkResultDict={}
+ count,minPvalue,plinkResultDict =self.getPlinkResultDict(outputFileName=plinkOutputFileName,thresholdPvalue=self.pValue,ChrOrderIdNameDict=ChrOrderIdNameDict)
+
+ # if can not find results which are matched with assigned p-value, system info will show up
+ if count >0:
+
+ #for genome association report table
+ reportTable=""
+ # sortable table object
+ resultstable,tblobj,bottomInfo = self.GenReportForPLINK(ChrNameOrderIdDict=ChrNameOrderIdDict, RISet=fd.RISet,plinkResultDict=plinkResultDict,thresholdPvalue=self.pValue,chrList=self.ChrList)
+
+ # creat object for result table for sort function
+ objfile = open('%s.obj' % (webqtlConfig.TMPDIR+filename), 'wb')
+ cPickle.dump(tblobj, objfile)
+ objfile.close()
+
+ sortby = ("Index", "up")
+ reportTable =HT.Div(webqtlUtil.genTableObj(tblobj=tblobj, file=filename, sortby=sortby, tableID = "sortable", addIndex = "0"), Id="sortable")
+
+ descriptionTable = HT.TableLite(border=0, cellpadding=0, cellspacing=0)
+ descriptionTable.append(HT.TR(HT.TD(reportTable, colspan=3)))
+ descriptionTable.append(HT.TR(HT.TD(HT.BR(),HT.BR())))
+ descriptionTable.append(bottomInfo)
+
+ # get each chr's length
+ self.ChrLengthMbList = map(lambda x: x/1000000.0, self.ChrLengthMbList) # change unit from bp to mb
+ self.ChrLengthMbSum = reduce(lambda x, y:x+y, self.ChrLengthMbList, 0.0)# get total length of all chrs
+ if self.ChrLengthMbList:
+ self.GraphInterval = self.ChrLengthMbSum/(len(self.ChrLengthMbList)*12) #Empirical Mb interval
+ else:
+ self.GraphInterval = 1
+
+ # for human data, there's no CM value
+ self.ChrLengthCMList = []
+ self.ChrLengthCMSum = 0
+
+ # begin: common part with human data
+ intCanvas = pid.PILCanvas(size=(self.graphWidth,self.graphHeight))
+ gifmap = self.plotIntMappingForPLINK(fd, intCanvas, startMb = self.startMb, endMb = self.endMb, plinkResultDict=plinkResultDict)
+
+ intCanvas.save(os.path.join(webqtlConfig.IMGDIR, filename), format='png')
+ intImg=HT.Image('/image/'+filename+'.png', border=0, usemap='#WebQTLImageMap')
+
+ TD_LR = HT.TR(HT.TD(HT.Blockquote(gifmap,intImg, HT.P()), bgColor='#eeeeee', height = 200))
+ self.dict['body'] = str(headerdiv)+str(TD_LR)+str(resultstable)+str(HT.TR(HT.TD(descriptionTable)))
+
+ else:
+ heading = "Genome Association"
+ detail = ['There is no association with marker that meets this criteria. Please provide a less stringend threshold. The minimun p-value is %s.'%minPvalue]
+ self.error(heading=heading,detail=detail)
+ return
+
+ elif self.mappingMethodId == '1': # QTLreaper result
+ if not fd.genotype:
+ fd.readData()
+
+ fd.parentsf14regression = fd.formdata.getvalue('parentsf14regression')
+ weightedRegression = fd.formdata.getvalue('applyVarianceSE')
+
+ if fd.parentsf14regression and fd.genotype_2:
+ _genotype = fd.genotype_2
+ else:
+ _genotype = fd.genotype_1
+
+ _strains, _vals, _vars, N = fd.informativeStrains(_genotype.prgy, weightedRegression)
+
+ if fd.identification:
+ heading2 = HT.Paragraph('Trait ID: %s' % fd.identification)
+ heading2.__setattr__("class","subtitle")
+ self.dict['title'] = '%s: Genome Association' % fd.identification
+ else:
+ heading2 = ""
+ self.dict['title'] = 'Genome Association'
+
+ if fd.traitInfo:
+ symbol,chromosome,MB = string.split(fd.traitInfo,'\t')
+ heading3 = HT.Paragraph('[ ',HT.Strong(HT.Italic('%s' % symbol,id="green")),' on Chr %s @ %s Mb ]' % (chromosome,MB))
+ else:
+ heading3 = ""
+
+ if N < webqtlConfig.KMININFORMATIVE:
+ heading = "Genome Association"
+ detail = ['Fewer than %d strain data were entered for %s data set. No mapping attempted.' % (webqtlConfig.KMININFORMATIVE, fd.RISet)]
+ self.error(heading=heading,detail=detail)
+ return
+ else:
+ heading = HT.Paragraph('Trait Data Entered for %s Set' % fd.RISet)
+ heading.__setattr__("class","title")
+
+ datadiv = HT.TD(heading, heading2,heading3, width='45%',valign='top', align='left', bgColor='#eeeeee')
+ resultstable,tblobj,bottomInfo = self.GenReport(ChrNameOrderIdDict,fd, _genotype, _strains, _vals, _vars)
+ #resultstable = self.GenReport(fd, _genotype, _strains, _vals, _vars)
+
+ # creat object for result table for sort function
+ objfile = open('%s.obj' % (webqtlConfig.TMPDIR+filename), 'wb')
+ cPickle.dump(tblobj, objfile)
+ objfile.close()
+
+ sortby = ("Index", "up")
+ reportTable =HT.Div(webqtlUtil.genTableObj(tblobj=tblobj, file=filename, sortby=sortby, tableID = "sortable", addIndex = "0"), Id="sortable")
+
+ descriptionTable = HT.TableLite(border=0, cellpadding=0, cellspacing=0)
+ descriptionTable.append(HT.TR(HT.TD(reportTable, colspan=3)))
+ descriptionTable.append(HT.TR(HT.TD(HT.BR(),HT.BR())))
+ descriptionTable.append(bottomInfo)
+
+ self.traitList=_vals
+
+ ##########################plot#######################
+
+ ################################################################
+ # Generate Chr list and Retrieve Length Information
+ ################################################################
+ self.genotype= _genotype
+ self.ChrList = [("All", -1)]
+
+ for i, indChr in enumerate(self.genotype):
+ self.ChrList.append((indChr.name, i))
+
+ self.cursor.execute("""
+ Select
+ Length from Chr_Length, InbredSet
+ where
+ Chr_Length.SpeciesId = InbredSet.SpeciesId AND
+ InbredSet.Name = '%s' AND
+ Chr_Length.Name in (%s)
+ Order by
+ OrderId
+ """ % (fd.RISet, string.join(map(lambda X: "'%s'" % X[0], self.ChrList[1:]), ", ")))
+
+ self.ChrLengthMbList = self.cursor.fetchall()
+ self.ChrLengthMbList = map(lambda x: x[0]/1000000.0, self.ChrLengthMbList)
+ self.ChrLengthMbSum = reduce(lambda x, y:x+y, self.ChrLengthMbList, 0.0)
+ if self.ChrLengthMbList:
+ self.MbGraphInterval = self.ChrLengthMbSum/(len(self.ChrLengthMbList)*12) #Empirical Mb interval
+ else:
+ self.MbGraphInterval = 1
+
+ self.ChrLengthCMList = []
+ for i, _chr in enumerate(self.genotype):
+ self.ChrLengthCMList.append(_chr[-1].cM - _chr[0].cM)
+ self.ChrLengthCMSum = reduce(lambda x, y:x+y, self.ChrLengthCMList, 0.0)# used for calculate plot scale
+
+ self.GraphInterval = self.MbGraphInterval #Mb
+
+ # begin: common part with human data
+ intCanvas = pid.PILCanvas(size=(self.graphWidth,self.graphHeight))
+ gifmap = self.plotIntMapping(fd, intCanvas, startMb = self.startMb, endMb = self.endMb, showLocusForm= "")
+ filename= webqtlUtil.genRandStr("Itvl_")
+ intCanvas.save(os.path.join(webqtlConfig.IMGDIR, filename), format='png')
+ intImg=HT.Image('/image/'+filename+'.png', border=0, usemap='#WebQTLImageMap')
+
+ ################################################################
+ # footnote goes here
+ ################################################################
+ btminfo = HT.Paragraph(Id="smallsize") #Small('More information about this graph is available here.')
+
+ if (self.additiveChecked):
+ btminfo.append(HT.BR(), 'A positive additive coefficient (', HT.Font('green', color='green'), ' line) indicates that %s alleles increase trait values. In contrast, a negative additive coefficient (' % fd.ppolar, HT.Font('red', color='red'), ' line) indicates that %s alleles increase trait values.' % fd.mpolar)
+
+
+ TD_LR = HT.TR(HT.TD(HT.Blockquote(gifmap,intImg, HT.P()), bgColor='#eeeeee', height = 200))
+
+ self.dict['body'] = str(datadiv)+str(TD_LR)+str(resultstable)+str(HT.TR(HT.TD(descriptionTable)))
+
+ # end: common part with human data
+
+ else:
+ pass
+
+
+ # add by NL 10-2-2011
+ def initializeParameters(self, fd):
+ """
+ Initializes all of the MarkerRegressionPage class parameters,
+ acquiring most values from the formdata (fd)
+ """
+ ###################################
+ # manhattam plot parameters
+ ###################################
+
+ self.graphHeight = 600
+ self.graphWidth = 1280
+ self.plotScale = 'physic'
+ self.selectedChr = -1
+ self.GRAPH_BACK_DARK_COLOR = pid.HexColor(0xF1F1F9)
+ self.GRAPH_BACK_LIGHT_COLOR = pid.HexColor(0xFBFBFF)
+ self.LRS_COLOR = pid.HexColor(0x0000FF)
+ self.LRS_LOD ='LRS'
+ self.lrsMax = float(fd.formdata.getvalue('lrsMax', 0))
+ self.startMb = fd.formdata.getvalue('startMb', "-1")
+ self.endMb = fd.formdata.getvalue('endMb', "-1")
+ self.mappingMethodId = fd.formdata.getvalue('mappingMethodId', "0")
+ self.permChecked=True
+ self.multipleInterval=False
+ self.SIGNIFICANT_WIDTH = 5
+ self.SUGGESTIVE_WIDTH = 5
+ self.SIGNIFICANT_COLOR = pid.HexColor(0xEBC7C7)
+ self.SUGGESTIVE_COLOR = pid.gainsboro
+ self.colorCollection = [self.LRS_COLOR]
+ self.additiveChecked= True
+ self.ADDITIVE_COLOR_POSITIVE = pid.green
+ self.legendChecked =False
+ self.pValue=float(fd.formdata.getvalue('pValue',-1))
+
+ # allow user to input p-value greater than 1,
+ # in this case, the value will be treated as -lgP value. so the input value needs to be transferred to power of 10 format
+ if self.pValue >1:
+ self.pValue =10**-(self.pValue)
+
+ try:
+ self.startMb = float(self.startMb)
+ self.endMb = float(self.endMb)
+ if self.startMb > self.endMb:
+ temp = self.startMb
+ self.startMb = self.endMb
+ self.endMb = temp
+ #minimal distance 10bp
+ if self.endMb - self.startMb < 0.00001:
+ self.endMb = self.startMb + 0.00001
+ except:
+ self.startMb = self.endMb = -1
+
+ def GenReportForPLINK(self, ChrNameOrderIdDict={},RISet='',plinkResultDict= {},thresholdPvalue=-1,chrList=[]):
+
+ 'Create an HTML division which reports any loci which are significantly associated with the submitted trait data.'
+ #########################################
+ # Genome Association report
+ #########################################
+ locusFormName = webqtlUtil.genRandStr("fm_")
+ locusForm = HT.Form(cgi = os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), \
+ enctype='multipart/form-data', name=locusFormName, submit=HT.Input(type='hidden'))
+ hddn = {'FormID':'showDatabase','ProbeSetID':'_','database':RISet+"Geno",'CellID':'_', \
+ 'RISet':RISet, 'incparentsf1':'on'}
+ for key in hddn.keys():
+ locusForm.append(HT.Input(name=key, value=hddn[key], type='hidden'))
+
+ regressionHeading = HT.Paragraph('Genome Association Report')
+ regressionHeading.__setattr__("class","title")
+
+ filename= webqtlUtil.genRandStr("GenomeAsscociation_")
+ fpText = open('%s.txt' % (webqtlConfig.TMPDIR+filename), 'wb')
+ fpText.write('The loci meet the criteria of P-Value <= %3.6f.\n'%thresholdPvalue)
+ pValueInfo =HT.Paragraph('The loci meet the criteria of P-Value <= %3.6f.\n'%thresholdPvalue)
+
+ textUrl = HT.Href(text = 'Download', url= '/tmp/'+filename+'.txt', target = "_blank", Class='fs12 fwn')
+ bottomInfo = HT.TR(HT.TD(HT.Paragraph(textUrl, ' result in tab-delimited text format.', HT.BR(), HT.BR(),Class="fs12 fwn"), colspan=3))
+
+ tblobj={} # build dict for genTableObj function; keys include header and body
+ tblobj_header = [] # value of key 'header'
+ tblobj_body=[] # value of key 'body'
+ reportHeaderRow=[] # header row list for tblobj_header (html part)
+ headerList=['Index','SNP Name','Chr','Mb','-log(P)']
+ headerStyle="fs14 fwb ffl b1 cw cbrb" # style of the header
+ cellColorStyle = "fs13 b1 fwn c222" # style of the cells
+
+ if headerList:
+ for ncol, item in enumerate(headerList):
+ reportHeaderRow.append(THCell(HT.TD(item, Class=headerStyle, valign='bottom',nowrap='ON'),text=item, idx=ncol))
+ #download file for table headers' names
+ fpText.write('SNP_Name\tChromosome\tMb\t-log(P)\n')
+
+ tblobj_header.append(reportHeaderRow)
+ tblobj['header']=tblobj_header
+
+ index=1
+ for chr in chrList:
+
+ if plinkResultDict.has_key(chr):
+ if chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[chr]
+ else:
+ chrOrderId=chr
+
+ valueList=plinkResultDict[chr]
+
+ for value in valueList:
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ snpName=value[0]
+ bp=value[1]
+ mb=int(bp)/1000000.0
+
+ try:
+ pValue =float(value[2])
+ except:
+ pValue =1
+ formattedPvalue = -math.log10(pValue)
+
+ formattedPvalue = webqtlUtil.SciFloat(formattedPvalue)
+ dbSnprs=snpName.replace('rs','')
+ SnpHref = HT.Href(text=snpName, url="http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=%s"%dbSnprs, target="_blank")
+
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(HT.TD(SnpHref, Class=cellColorStyle,nowrap='ON'),snpName, snpName))
+ reportBodyRow.append(TDCell(HT.TD(chr, Class=cellColorStyle, align="center",nowrap='ON'),chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%mb, Class=cellColorStyle, align="center",nowrap='ON'),mb, mb))
+ reportBodyRow.append(TDCell(HT.TD(formattedPvalue, Class=cellColorStyle, align="center",nowrap='ON'),formattedPvalue, float(formattedPvalue)))
+
+ fpText.write('%s\t%s\t%3.6f\t%s\n' % (snpName, str(chr), mb, formattedPvalue))
+ index+=1
+
+ tblobj_body.append(reportBodyRow)
+
+ tblobj['body']=tblobj_body
+ rv=HT.TR(HT.TD(regressionHeading,pValueInfo, locusForm, HT.P(), width='55%',valign='top', align='left',bgColor='#eeeeee'))
+
+ return rv, tblobj,bottomInfo
+
+
+ def GenReport(self, ChrNameOrderIdDict,fd, _genotype, _strains, _vals, _vars= []):
+ 'Create an HTML division which reports any loci which are significantly associated with the submitted trait data.'
+ #calculate QTL for each trait
+ self.qtlresults = []
+ if webqtlUtil.ListNotNull(_vars):
+ qtlresults = _genotype.regression(strains = _strains, trait = _vals, variance = _vars)
+ LRSArray = _genotype.permutation(strains = _strains, trait = _vals, variance = _vars, nperm=fd.nperm)
+ else:
+ qtlresults = _genotype.regression(strains = _strains, trait = _vals)
+ LRSArray = _genotype.permutation(strains = _strains, trait = _vals,nperm=fd.nperm)
+
+ self.qtlresults.append(qtlresults)
+
+ filename= webqtlUtil.genRandStr("GenomeAsscociation_")
+
+ # set suggestive, significant and highly significant LRS
+ if fd.suggestive == None:
+ fd.suggestive = LRSArray[int(fd.nperm*0.37-1)]
+ else:
+ fd.suggestive = float(fd.suggestive)
+ if fd.significance == None:
+ fd.significance = LRSArray[int(fd.nperm*0.95-1)]
+ else:
+ fd.significance = float(fd.significance)
+
+ self.significance =fd.significance
+ self.suggestive = fd.suggestive
+ self.highlysignificant = LRSArray[int(fd.nperm*0.99-1)]
+ _dispAllLRS = 0
+ if fd.formdata.getvalue('displayAllLRS'):
+ _dispAllLRS = 1
+ qtlresults2 = []
+ if _dispAllLRS:
+ filtered = qtlresults[:]
+ else:
+ filtered = filter(lambda x, y=fd.suggestive: x.lrs > y, qtlresults)
+ if len(filtered) == 0:
+ qtlresults2 = qtlresults[:]
+ qtlresults2.sort()
+ filtered = qtlresults2[-10:]
+
+ #########################################
+ # Permutation Graph
+ #########################################
+ myCanvas = pid.PILCanvas(size=(400,300))
+ #plotBar(myCanvas,10,10,390,290,LRSArray,XLabel='LRS',YLabel='Frequency',title=' Histogram of Permutation Test',identification=fd.identification)
+ Plot.plotBar(myCanvas, LRSArray,XLabel='LRS',YLabel='Frequency',title=' Histogram of Permutation Test')
+ filename= webqtlUtil.genRandStr("Reg_")
+ myCanvas.save(webqtlConfig.IMGDIR+filename, format='gif')
+ img=HT.Image('/image/'+filename+'.gif',border=0,alt='Histogram of Permutation Test')
+
+ if fd.suggestive == None:
+ fd.suggestive = LRSArray[int(fd.nperm*0.37-1)]
+ else:
+ fd.suggestive = float(fd.suggestive)
+ if fd.significance == None:
+ fd.significance = LRSArray[int(fd.nperm*0.95-1)]
+ else:
+ fd.significance = float(fd.significance)
+
+ permutationHeading = HT.Paragraph('Histogram of Permutation Test')
+ permutationHeading.__setattr__("class","title")
+
+ permutation = HT.TableLite()
+ permutation.append(HT.TR(HT.TD(img)))
+
+
+ #########################################
+ # Genome Association report
+ #########################################
+ locusFormName = webqtlUtil.genRandStr("fm_")
+ locusForm = HT.Form(cgi = os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), \
+ enctype='multipart/form-data', name=locusFormName, submit=HT.Input(type='hidden'))
+ hddn = {'FormID':'showDatabase','ProbeSetID':'_','database':fd.RISet+"Geno",'CellID':'_', \
+ 'RISet':fd.RISet, 'incparentsf1':'on'}
+ for key in hddn.keys():
+ locusForm.append(HT.Input(name=key, value=hddn[key], type='hidden'))
+
+ regressionHeading = HT.Paragraph('Genome Association Report')
+ regressionHeading.__setattr__("class","title")
+ # report is the info part above report table
+ if qtlresults2 != []:
+ report = HT.Blockquote(HT.Font('No association ',color="#FF0000"),HT.Font('with a likelihood ratio statistic greater than %3.1f was found. Here are the top 10 LRSs.' % fd.suggestive,color="#000000"))
+ else:
+ report = HT.Blockquote('The following loci in the %s data set have associations with the above trait data.\n' % fd.RISet, HT.P())
+ report.__setattr__("class","normalsize")
+
+ fpText = open('%s.txt' % (webqtlConfig.TMPDIR+filename), 'wb')
+ fpText.write('Suggestive LRS =%3.2f\n'%self.suggestive)
+ fpText.write('Significant LRS =%3.2f\n'%self.significance)
+ fpText.write('Highly Significant LRS =%3.2f\n'%self.highlysignificant)
+ LRSInfo =HT.Paragraph(' Suggestive LRS =%3.2f\n'%fd.suggestive, HT.BR(), ' Significant LRS =%3.2f\n'%fd.significance,HT.BR(),' Highly Significant LRS =%3.2f\n' % self.highlysignificant)
+
+ textUrl = HT.Href(text = 'Download', url= '/tmp/'+filename+'.txt', target = "_blank", Class='fs12 fwn')
+
+ bottomInfo = HT.TR(HT.TD(HT.Paragraph(textUrl, ' result in tab-delimited text format.', HT.BR(), HT.BR(),'LRS values marked with',HT.Font(' * ',color="red"), 'are greater than the significance threshold (specified by you or by permutation test). ' , HT.BR(), HT.BR(), HT.Strong('Additive Effect'), ' is half the difference in the mean phenotype of all cases that are homozygous for one parental allel at this marker minus the mean of all cases that are homozygous for the other parental allele at this marker. ','In the case of %s strains, for example,' % fd.RISet,' A positive additive effect indicates that %s alleles increase trait values. Negative additive effect indicates that %s alleles increase trait values.'% (fd.ppolar,fd.mpolar),Class="fs12 fwn")))
+
+ tblobj={} # build dict for genTableObj function; keys include header and body
+ tblobj_header = [] # value of key 'header'
+ tblobj_body=[] # value of key 'body'
+ reportHeaderRow=[] # header row list for tblobj_header (html part)
+ headerStyle="fs14 fwb ffl b1 cw cbrb" # style of the header
+ cellColorStyle = "fs13 b1 fwn c222" # style of the cells
+
+ headerList=['Index','LRS','Chr','Mb','Locus','Additive Effect']
+ for ncol, item in enumerate(headerList):
+ reportHeaderRow.append(THCell(HT.TD(item, Class=headerStyle, valign='bottom',nowrap='ON'),text=item, idx=ncol))
+
+ if fd.genotype.type == 'intercross':
+ ncol =len(headerList)
+ reportHeaderRow.append(THCell(HT.TD('Dominance Effect', Class=headerStyle, valign='bottom',nowrap='ON'),text='Dominance Effect', idx=ncol))
+
+ #download file for table headers' names
+ fpText.write('LRS\tChromosome\tMb\tLocus\tAdditive Effect\tDominance Effect\n')
+
+ index=1
+ for ii in filtered:
+ #add by NL 06-20-2011: set LRS to 460 when LRS is infinite,
+ if ii.lrs==float('inf') or ii.lrs>webqtlConfig.MAXLRS:
+ LRS=webqtlConfig.MAXLRS #maximum LRS value
+ else:
+ LRS=ii.lrs
+
+ if LRS > fd.significance:
+ lrs = HT.TD(HT.Font('%3.3f*' % LRS, color='#FF0000'),Class=cellColorStyle)
+ else:
+ lrs = HT.TD('%3.3f' % LRS,Class=cellColorStyle)
+
+ if ii.locus.chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[ii.locus.chr]
+ else:
+ chrOrderId=ii.locus.chr
+
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(lrs,LRS, LRS))
+ reportBodyRow.append(TDCell(HT.TD(ii.locus.chr, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%ii.locus.Mb, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.Mb, ii.locus.Mb))
+ reportBodyRow.append(TDCell(HT.TD(HT.Href(text=ii.locus.name, url = "javascript:showTrait('%s','%s');" % (locusFormName, ii.locus.name), Class='normalsize'), Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.name, ii.locus.name))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.additive, Class=cellColorStyle, align="center",nowrap='ON'),ii.additive, ii.additive))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.dominance, Class=cellColorStyle, align="center",nowrap='ON'),ii.dominance, ii.dominance))
+
+ fpText.write('%2.3f\t%s\t%3.6f\t%s\t%2.3f\t%2.3f\n' % (LRS, ii.locus.chr, ii.locus.Mb, ii.locus.name, ii.additive, ii.dominance))
+ index+=1
+ tblobj_body.append(reportBodyRow)
+ else:
+ #download file for table headers' names
+ fpText.write('LRS\tChromosome\tMb\tLocus\tAdditive Effect\n')
+
+ index=1
+ for ii in filtered:
+ #add by NL 06-20-2011: set LRS to 460 when LRS is infinite,
+ if ii.lrs==float('inf') or ii.lrs>webqtlConfig.MAXLRS:
+ LRS=webqtlConfig.MAXLRS #maximum LRS value
+ else:
+ LRS=ii.lrs
+
+ if LRS > fd.significance:
+ lrs = HT.TD(HT.Font('%3.3f*' % LRS, color='#FF0000'),Class=cellColorStyle)
+ else:
+ lrs = HT.TD('%3.3f' % LRS,Class=cellColorStyle)
+
+ if ii.locus.chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[ii.locus.chr]
+ else:
+ chrOrderId=ii.locus.chr
+
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(lrs,LRS, LRS))
+ reportBodyRow.append(TDCell(HT.TD(ii.locus.chr, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%ii.locus.Mb, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.Mb, ii.locus.Mb))
+ reportBodyRow.append(TDCell(HT.TD(HT.Href(text=ii.locus.name, url = "javascript:showTrait('%s','%s');" % (locusFormName, ii.locus.name), Class='normalsize'), Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.name, ii.locus.name))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.additive, Class=cellColorStyle, align="center",nowrap='ON'),ii.additive, ii.additive))
+
+ fpText.write('%2.3f\t%s\t%3.6f\t%s\t%2.3f\n' % (LRS, ii.locus.chr, ii.locus.Mb, ii.locus.name, ii.additive))
+ index+=1
+ tblobj_body.append(reportBodyRow)
+
+ tblobj_header.append(reportHeaderRow)
+ tblobj['header']=tblobj_header
+ tblobj['body']=tblobj_body
+
+ rv=HT.TD(regressionHeading,LRSInfo,report, locusForm, HT.P(),width='55%',valign='top', align='left', bgColor='#eeeeee')
+ if fd.genotype.type == 'intercross':
+ bottomInfo.append(HT.BR(), HT.BR(), HT.Strong('Dominance Effect'),' is the difference between the mean trait value of cases heterozygous at a marker and the average mean for the two groups homozygous at this marker: e.g., BD - (BB+DD)/2]. A positive dominance effect indicates that the average phenotype of BD heterozygotes exceeds the mean of BB and DD homozygotes. No dominance deviation can be computed for a set of recombinant inbred strains or for a backcross.')
+ return rv,tblobj,bottomInfo
+
+ return rv,tblobj,bottomInfo
+
+ def plotIntMappingForPLINK(self, fd, canvas, offset= (80, 120, 20, 80), zoom = 1, startMb = None, endMb = None, showLocusForm = "",plinkResultDict={}):
+ #calculating margins
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ xLeftOffset = int(xLeftOffset*fontZoom)
+ xRightOffset = int(xRightOffset*fontZoom)
+ yBottomOffset = int(yBottomOffset*fontZoom)
+
+ cWidth = canvas.size[0]
+ cHeight = canvas.size[1]
+ plotWidth = cWidth - xLeftOffset - xRightOffset
+ plotHeight = cHeight - yTopOffset - yBottomOffset
+ startPixelX = xLeftOffset
+ endPixelX = (xLeftOffset + plotWidth)
+
+ #Drawing Area Height
+ drawAreaHeight = plotHeight
+ if self.plotScale == 'physic' and self.selectedChr > -1: # for single chr
+ drawAreaHeight -= self.ENSEMBL_BAND_HEIGHT + self.UCSC_BAND_HEIGHT+ self.WEBQTL_BAND_HEIGHT + 3*self.BAND_SPACING+ 10*zoom
+ if self.geneChecked:
+ drawAreaHeight -= self.NUM_GENE_ROWS*self.EACH_GENE_HEIGHT + 3*self.BAND_SPACING + 10*zoom
+ else:
+ if self.selectedChr > -1:
+ drawAreaHeight -= 20
+ else:# for all chrs
+ drawAreaHeight -= 30
+
+ #Image map
+ gifmap = HT.Map(name='WebQTLImageMap')
+
+ newoffset = (xLeftOffset, xRightOffset, yTopOffset, yBottomOffset)
+ # Draw the alternating-color background first and get plotXScale
+ plotXScale = self.drawGraphBackgroundForPLINK(canvas, gifmap, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb,plinkResultDict=plinkResultDict)
+
+ # Draw X axis
+ self.drawXAxisForPLINK(fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+ # Draw manhattam plot
+ self.drawManhattanPlotForPLINK(canvas, drawAreaHeight, gifmap, plotXScale, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb,plinkResultDict=plinkResultDict,thresholdPvalue=self.pValue)
+
+ return gifmap
+
+
+ def plotIntMapping(self, fd, canvas, offset= (80, 120, 20, 80), zoom = 1, startMb = None, endMb = None, showLocusForm = ""):
+ #calculating margins
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ xLeftOffset = int(xLeftOffset*fontZoom)
+ xRightOffset = int(xRightOffset*fontZoom)
+ yBottomOffset = int(yBottomOffset*fontZoom)
+
+ cWidth = canvas.size[0]
+ cHeight = canvas.size[1]
+ plotWidth = cWidth - xLeftOffset - xRightOffset
+ plotHeight = cHeight - yTopOffset - yBottomOffset
+ startPixelX = xLeftOffset
+ endPixelX = (xLeftOffset + plotWidth)
+
+ #Drawing Area Height
+ drawAreaHeight = plotHeight
+ if self.plotScale == 'physic' and self.selectedChr > -1: # for single chr
+ drawAreaHeight -= self.ENSEMBL_BAND_HEIGHT + self.UCSC_BAND_HEIGHT+ self.WEBQTL_BAND_HEIGHT + 3*self.BAND_SPACING+ 10*zoom
+ if self.geneChecked:
+ drawAreaHeight -= self.NUM_GENE_ROWS*self.EACH_GENE_HEIGHT + 3*self.BAND_SPACING + 10*zoom
+ else:# for all chrs
+ if self.selectedChr > -1:
+ drawAreaHeight -= 20
+ else:
+ drawAreaHeight -= 30
+
+ #Image map
+ gifmap = HT.Map(name='WebQTLImageMap')
+
+ newoffset = (xLeftOffset, xRightOffset, yTopOffset, yBottomOffset)
+ # Draw the alternating-color background first and get plotXScale
+ plotXScale = self.drawGraphBackground(canvas, gifmap, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+
+ # Draw X axis
+ self.drawXAxis(fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+ # Draw QTL curve
+ self.drawQTL(canvas, drawAreaHeight, gifmap, plotXScale, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+
+ #draw legend
+ if self.multipleInterval:
+ self.drawMultiTraitName(fd, canvas, gifmap, showLocusForm, offset=newoffset)
+ elif self.legendChecked:
+ self.drawLegendPanel(fd, canvas, offset=newoffset)
+ else:
+ pass
+
+ #draw position, no need to use a separate function
+ if fd.genotype.Mbmap:
+ self.drawProbeSetPosition(canvas, plotXScale, offset=newoffset)
+
+ return gifmap
+
+
+ # functions for manhattam plot of markers
+ def drawManhattanPlotForPLINK(self, canvas, drawAreaHeight, gifmap, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None,plinkResultDict={},thresholdPvalue=-1):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ # INTERCROSS = (self.genotype.type=="intercross")
+ INTERCROSS ='' #??????
+
+ ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+ GraphInterval=self.GraphInterval
+ pvalueHeightThresh = drawAreaHeight - 80 #ZS: Otherwise the plot gets very close to the chromosome labels
+
+ #draw the pvalue scale
+ #We first determine whether or not we are using a sliding scale.
+ #If so, we need to compute the maximum pvalue value to determine where the max y-value should be, and call this pvalueMax.
+ #pvalueTop is then defined to be above the pvalueMax by enough to add one additional pvalueScale increment.
+ #if we are using a set-scale, then we set pvalueTop to be the user's value, and pvalueMax doesn't matter.
+
+ # for human data we use p value instead of lrs
+ pValueList=[]
+ for key in plinkResultDict:
+ valueList = plinkResultDict[key]
+ for item in valueList:
+ pValue = item[-1]
+ pValueList.append(pValue)
+
+ formattedPValueList=[]
+ for pValue in pValueList:
+ try:
+ pValue=float(pValue)
+ except:
+ pValue =1
+ formattedpValue = -math.log10(pValue)
+ formattedPValueList.append(formattedpValue)
+
+ #sliding scale
+ pvalueMax = max(formattedPValueList)
+ #pvalueMax =pvalueMax +1
+ # no permutation result for plink func: GenReport()
+ pvalueMin = int(-math.log10(thresholdPvalue))
+
+ if pvalueMax> 100:
+ pvalueScale = 20.0
+ elif pvalueMax > 20:
+ pvalueScale = 5.0
+ elif pvalueMax > 7.5:
+ pvalueScale = 2.5
+ else:
+ pvalueScale = 1.0
+
+ # the base line for x-axis is -log(thresholdPvalue)
+ pvalueAxisList = Plot.frange(pvalueMin, pvalueMax, pvalueScale)
+ #make sure the user's value appears on the y-axis
+ #ZS: There is no way to do this without making the position of the points not directly proportional to a given distance on the y-axis
+ #tempPvalueMax=round(pvalueMax)
+ tempPvalueMax = pvalueAxisList[len(pvalueAxisList)-1] + pvalueScale
+ pvalueAxisList.append(tempPvalueMax)
+
+ #ZS: I don't understand this; the if statement will be true for any number that isn't exactly X.5.
+ #if abs(tempPvalueMax-pvalueMax) <0.5:
+ # tempPvalueMax=tempPvalueMax+1
+ # pvalueAxisList.append(tempPvalueMax)
+
+ #draw the "pvalue" string to the left of the axis
+ pvalueScaleFont=pid.Font(ttf="verdana", size=14*fontZoom, bold=0)
+ pvalueLODFont=pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ yZero = yTopOffset + plotHeight
+
+ #yAxis label display area
+ yAxis_label ='-log(P)'
+ canvas.drawString(yAxis_label, xLeftOffset - canvas.stringWidth("999.99", font=pvalueScaleFont) - 10*zoom, \
+ yZero - 150, font=pvalueLODFont, color=pid.black, angle=90)
+
+ for i,item in enumerate(pvalueAxisList):
+ ypvalue = yZero - (float(i)/float(len(pvalueAxisList) - 1)) * pvalueHeightThresh
+ canvas.drawLine(xLeftOffset, ypvalue, xLeftOffset - 4, ypvalue, color=self.LRS_COLOR, width=1*zoom)
+ scaleStr = "%2.1f" % item
+ #added by NL 6-24-2011:Y-axis scale display
+ canvas.drawString(scaleStr, xLeftOffset-4-canvas.stringWidth(scaleStr, font=pvalueScaleFont)-5, ypvalue+3, font=pvalueScaleFont, color=self.LRS_COLOR)
+
+ ChrList=self.ChrList
+ startPosX = xLeftOffset
+
+ for i, chr in enumerate(ChrList):
+
+ if plinkResultDict.has_key(chr):
+ plinkresultList = plinkResultDict[chr]
+
+ m = 0
+ #add by NL 06-24-2011: for mahanttam plot
+ symbolFont = pid.Font(ttf="fnt_bs", size=5,bold=0)
+ # color for point in each chr
+ chrCount=len(ChrList)
+ chrColorDict =self.getColorForMarker(chrCount=chrCount,flag=1)
+ for j, item in enumerate(plinkresultList):
+ try :
+ mb=float(item[1])/1000000.0
+ except:
+ mb=0
+
+ try :
+ pvalue =float(item[-1])
+ except:
+ pvalue =1
+
+ try:
+ snpName = item[0]
+ except:
+ snpName=''
+
+ formattedPvalue = -math.log10(pvalue)
+
+ Xc = startPosX + (mb-startMb)*plotXScale
+ Yc = yZero - (formattedPvalue-pvalueMin)*pvalueHeightThresh/(tempPvalueMax - pvalueMin)
+ canvas.drawString("5", Xc-canvas.stringWidth("5",font=symbolFont)/2+1,Yc+2,color=chrColorDict[i], font=symbolFont)
+ m += 1
+
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset, yTopOffset, color=self.LRS_COLOR, width=1*zoom) #the blue line running up the y axis
+
+ def drawQTL(self, canvas, drawAreaHeight, gifmap, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ INTERCROSS = (self.genotype.type=="intercross")
+
+ ChrLengthDistList = self.ChrLengthMbList
+ GraphInterval=self.GraphInterval
+ LRSHeightThresh = drawAreaHeight
+ AdditiveHeightThresh = drawAreaHeight/2
+ DominanceHeightThresh = drawAreaHeight/2
+
+ #draw the LRS scale
+ #We first determine whether or not we are using a sliding scale.
+ #If so, we need to compute the maximum LRS value to determine where the max y-value should be, and call this LRSMax.
+ #LRSTop is then defined to be above the LRSMax by enough to add one additional LRSScale increment.
+ #if we are using a set-scale, then we set LRSTop to be the user's value, and LRSMax doesn't matter.
+
+ if self.LRS_LOD == 'LOD':
+ lodm = self.LODFACTOR
+ else:
+ lodm = 1.0
+
+ if self.lrsMax <= 0: #sliding scale
+ LRSMax = max(map(max, self.qtlresults)).lrs
+ #genotype trait will give infinite LRS
+ LRSMax = min(LRSMax, webqtlConfig.MAXLRS)
+ LRSMax = max(self.significance, LRSMax)
+ else:
+ LRSMax = self.lrsMax*lodm
+
+ if LRSMax/lodm > 100:
+ LRSScale = 20.0
+ elif LRSMax/lodm > 20:
+ LRSScale = 5.0
+ elif LRSMax/lodm > 7.5:
+ LRSScale = 2.5
+ else:
+ LRSScale = 1.0
+
+ LRSAxisList = Plot.frange(LRSScale, LRSMax/lodm, LRSScale)
+ #make sure the user's value appears on the y-axis
+ #update by NL 6-21-2011: round the LOD value to 100 when LRSMax is equal to 460
+ LRSAxisList.append(round(LRSMax/lodm))
+
+ #draw the "LRS" or "LOD" string to the left of the axis
+ LRSScaleFont=pid.Font(ttf="verdana", size=14*fontZoom, bold=0)
+ LRSLODFont=pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ yZero = yTopOffset + plotHeight
+
+ #yAxis label display area
+ canvas.drawString(self.LRS_LOD, xLeftOffset - canvas.stringWidth("999.99", font=LRSScaleFont) - 10*zoom, \
+ yZero - 150, font=LRSLODFont, color=pid.black, angle=90)
+
+ for item in LRSAxisList:
+ yLRS = yZero - (item*lodm/LRSMax) * LRSHeightThresh
+ canvas.drawLine(xLeftOffset, yLRS, xLeftOffset - 4, yLRS, color=self.LRS_COLOR, width=1*zoom)
+ scaleStr = "%2.1f" % item
+ #added by NL 6-24-2011:Y-axis scale display
+ canvas.drawString(scaleStr, xLeftOffset-4-canvas.stringWidth(scaleStr, font=LRSScaleFont)-5, yLRS+3, font=LRSScaleFont, color=self.LRS_COLOR)
+
+
+ #"Significant" and "Suggestive" Drawing Routine
+ # ======= Draw the thick lines for "Significant" and "Suggestive" ===== (crowell: I tried to make the SNPs draw over these lines, but piddle wouldn't have it...)
+ if self.permChecked and not self.multipleInterval:
+ significantY = yZero - self.significance*LRSHeightThresh/LRSMax
+ suggestiveY = yZero - self.suggestive*LRSHeightThresh/LRSMax
+
+
+ startPosX = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ rightEdge = int(startPosX + self.ChrLengthDistList[i]*plotXScale - self.SUGGESTIVE_WIDTH/1.5)
+ #added by NL 6-24-2011:draw suggestive line (grey one)
+ canvas.drawLine(startPosX+self.SUGGESTIVE_WIDTH/1.5, suggestiveY, rightEdge, suggestiveY, color=self.SUGGESTIVE_COLOR,
+ width=self.SUGGESTIVE_WIDTH*zoom, clipX=(xLeftOffset, xLeftOffset + plotWidth-2))
+ #added by NL 6-24-2011:draw significant line (pink one)
+ canvas.drawLine(startPosX+self.SUGGESTIVE_WIDTH/1.5, significantY, rightEdge, significantY, color=self.SIGNIFICANT_COLOR,
+ width=self.SIGNIFICANT_WIDTH*zoom, clipX=(xLeftOffset, xLeftOffset + plotWidth-2))
+ sugg_coords = "%d, %d, %d, %d" % (startPosX, suggestiveY-2, rightEdge + 2*zoom, suggestiveY+2)
+ sig_coords = "%d, %d, %d, %d" % (startPosX, significantY-2, rightEdge + 2*zoom, significantY+2)
+ if self.LRS_LOD == 'LRS':
+ sugg_title = "Suggestive LRS = %0.2f" % self.suggestive
+ sig_title = "Significant LRS = %0.2f" % self.significance
+ else:
+ sugg_title = "Suggestive LOD = %0.2f" % (self.suggestive/4.61)
+ sig_title = "Significant LOD = %0.2f" % (self.significance/4.61)
+ Areas1 = HT.Area(shape='rect',coords=sugg_coords,title=sugg_title)
+ Areas2 = HT.Area(shape='rect',coords=sig_coords,title=sig_title)
+ gifmap.areas.append(Areas1)
+ gifmap.areas.append(Areas2)
+
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+
+ if self.multipleInterval:
+ lrsEdgeWidth = 1
+ else:
+ additiveMax = max(map(lambda X : abs(X.additive), self.qtlresults[0]))
+ if INTERCROSS:
+ dominanceMax = max(map(lambda X : abs(X.dominance), self.qtlresults[0]))
+ else:
+ dominanceMax = -1
+ lrsEdgeWidth = 2
+ for i, qtlresult in enumerate(self.qtlresults):
+ m = 0
+ startPosX = xLeftOffset
+ thisLRSColor = self.colorCollection[i]
+
+ #add by NL 06-24-2011: for mahanttam plot
+ symbolFont = pid.Font(ttf="fnt_bs", size=5,bold=0)
+
+ for j, _chr in enumerate(self.genotype):
+ chrCount=len(self.genotype)
+ chrColorDict =self.getColorForMarker(chrCount=chrCount,flag=1)
+ LRSCoordXY = []
+ AdditiveCoordXY = []
+ DominanceCoordXY = []
+ for k, _locus in enumerate(_chr):
+ if self.plotScale == 'physic':
+ Xc = startPosX + (_locus.Mb-startMb)*plotXScale
+ else:
+ Xc = startPosX + (_locus.cM-_chr[0].cM)*plotXScale
+ # updated by NL 06-18-2011:
+ # fix the over limit LRS graph issue since genotype trait may give infinite LRS;
+ # for any lrs is over than 460(LRS max in this system), it will be reset to 460
+ if qtlresult[m].lrs> 460 or qtlresult[m].lrs=='inf':
+ Yc = yZero - webqtlConfig.MAXLRS*LRSHeightThresh/LRSMax
+ else:
+ Yc = yZero - qtlresult[m].lrs*LRSHeightThresh/LRSMax
+
+ LRSCoordXY.append((Xc, Yc))
+ #add by NL 06-24-2011: for mahanttam plot
+ #self.significance/4.61 consider chr and LOD
+ # significantY = yZero - self.significance*LRSHeightThresh/LRSMax
+ # if Yc >significantY:
+ # canvas.drawString(":", Xc-canvas.stringWidth(":",font=symbolFont)/2+1,Yc+2,color=pid.black, font=symbolFont)
+ # else:
+ # canvas.drawString(":", Xc-canvas.stringWidth(":",font=symbolFont)/2+1,Yc+2,color=pid.black, font=symbolFont)
+
+ # add by NL 06-27-2011: eliminate imputed value when locus name is equal to '-'
+ if (qtlresult[m].locus.name) and (qtlresult[m].locus.name!=' - '):
+ canvas.drawString("5", Xc-canvas.stringWidth("5",font=symbolFont)/2+1,Yc+2,color=chrColorDict[j], font=symbolFont)
+
+ if not self.multipleInterval and self.additiveChecked:
+ Yc = yZero - qtlresult[m].additive*AdditiveHeightThresh/additiveMax
+ AdditiveCoordXY.append((Xc, Yc))
+ if not self.multipleInterval and INTERCROSS and self.additiveChecked:
+ Yc = yZero - qtlresult[m].dominance*DominanceHeightThresh/dominanceMax
+ DominanceCoordXY.append((Xc, Yc))
+ m += 1
+
+ startPosX += (ChrLengthDistList[j]+GraphInterval)*plotXScale
+
+
+ ###draw additive scale
+ if not self.multipleInterval and self.additiveChecked:
+ additiveScaleFont=pid.Font(ttf="verdana",size=12*fontZoom,bold=0)
+ additiveScale = Plot.detScaleOld(0,additiveMax)
+ additiveStep = (additiveScale[1]-additiveScale[0])/additiveScale[2]
+ additiveAxisList = Plot.frange(0, additiveScale[1], additiveStep)
+ maxAdd = additiveScale[1]
+ addPlotScale = AdditiveHeightThresh/additiveMax
+
+ additiveAxisList.append(additiveScale[1])
+ for item in additiveAxisList:
+ additiveY = yZero - item*addPlotScale
+ canvas.drawLine(xLeftOffset + plotWidth,additiveY,xLeftOffset+4+ plotWidth,additiveY,color=self.ADDITIVE_COLOR_POSITIVE, width=1*zoom)
+ scaleStr = "%2.3f" % item
+ canvas.drawString(scaleStr,xLeftOffset + plotWidth +6,additiveY+5,font=additiveScaleFont,color=self.ADDITIVE_COLOR_POSITIVE)
+
+ canvas.drawLine(xLeftOffset+plotWidth,additiveY,xLeftOffset+plotWidth,yZero,color=self.ADDITIVE_COLOR_POSITIVE, width=1*zoom)
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset, yTopOffset, color=self.LRS_COLOR, width=1*zoom) #the blue line running up the y axis
+
+ def drawGraphBackgroundForPLINK(self, canvas, gifmap, offset= (80, 120, 80, 50), zoom = 1, startMb = None, endMb = None,plinkResultDict={} ):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #calculate plot scale
+ #XZ: all of these global variables should be passed from function signiture
+ ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+ GraphInterval=self.GraphInterval
+ ChrList =self.ChrList
+
+ #multiple chromosome view
+ plotXScale = plotWidth / ((len(ChrList)-1)*GraphInterval + drawRegionDistance)
+
+ startPosX = xLeftOffset
+ chrLabelFont=pid.Font(ttf="verdana",size=24*fontZoom,bold=0)
+
+ for i, _chr in enumerate(ChrList):
+
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+ # NL:resize chr width for drawing
+ if float(ChrLengthDistList[i])<90:
+ ChrLengthDistList[i]=90
+ #draw the shaded boxes and the sig/sug thick lines
+ canvas.drawRect(startPosX, yTopOffset, startPosX + ChrLengthDistList[i]*plotXScale, \
+ yTopOffset+plotHeight, edgeColor=pid.gainsboro,fillColor=theBackColor)
+
+ chrNameWidth = canvas.stringWidth(_chr, font=chrLabelFont)
+ chrStartPix = startPosX + (ChrLengthDistList[i]*plotXScale -chrNameWidth)/2
+ chrEndPix = startPosX + (ChrLengthDistList[i]*plotXScale +chrNameWidth)/2
+
+ canvas.drawString(_chr, chrStartPix, yTopOffset +20,font = chrLabelFont,color=pid.dimgray)
+ COORDS = "%d,%d,%d,%d" %(chrStartPix, yTopOffset, chrEndPix,yTopOffset +20)
+
+ #add by NL 09-03-2010
+ HREF = "javascript:changeView(%d,%s);" % (i,ChrLengthDistList)
+ Areas = HT.Area(shape='rect',coords=COORDS,href=HREF)
+ gifmap.areas.append(Areas)
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ return plotXScale
+
+
+ def drawGraphBackground(self, canvas, gifmap, offset= (80, 120, 80, 50), zoom = 1, startMb = None, endMb = None):
+ ##conditions
+ ##multiple Chromosome view
+ ##single Chromosome Physical
+ ##single Chromosome Genetic
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #calculate plot scale
+ if self.plotScale != 'physic':
+ self.ChrLengthDistList = self.ChrLengthCMList
+ drawRegionDistance = self.ChrLengthCMSum
+ else:
+ self.ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+
+ if self.selectedChr > -1: #single chromosome view
+ spacingAmt = plotWidth/13.5
+ i = 0
+ for startPix in Plot.frange(xLeftOffset, xLeftOffset+plotWidth, spacingAmt):
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+ i += 1
+ canvas.drawRect(startPix, yTopOffset, min(startPix+spacingAmt, xLeftOffset+plotWidth), \
+ yTopOffset+plotHeight, edgeColor=theBackColor, fillColor=theBackColor)
+
+ drawRegionDistance = self.ChrLengthDistList[self.selectedChr]
+ self.ChrLengthDistList = [drawRegionDistance]
+ if self.plotScale == 'physic':
+ plotXScale = plotWidth / (endMb-startMb)
+ else:
+ plotXScale = plotWidth / drawRegionDistance
+
+ else: #multiple chromosome view
+ plotXScale = plotWidth / ((len(self.genotype)-1)*self.GraphInterval + drawRegionDistance)
+
+ startPosX = xLeftOffset
+ chrLabelFont=pid.Font(ttf="verdana",size=24*fontZoom,bold=0)
+
+ for i, _chr in enumerate(self.genotype):
+
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+
+ #draw the shaded boxes and the sig/sug thick lines
+ canvas.drawRect(startPosX, yTopOffset, startPosX + self.ChrLengthDistList[i]*plotXScale, \
+ yTopOffset+plotHeight, edgeColor=pid.gainsboro,fillColor=theBackColor)
+
+ chrNameWidth = canvas.stringWidth(_chr.name, font=chrLabelFont)
+ chrStartPix = startPosX + (self.ChrLengthDistList[i]*plotXScale -chrNameWidth)/2
+ chrEndPix = startPosX + (self.ChrLengthDistList[i]*plotXScale +chrNameWidth)/2
+
+ canvas.drawString(_chr.name, chrStartPix, yTopOffset +20,font = chrLabelFont,color=pid.dimgray)
+ COORDS = "%d,%d,%d,%d" %(chrStartPix, yTopOffset, chrEndPix,yTopOffset +20)
+
+ #add by NL 09-03-2010
+ HREF = "javascript:changeView(%d,%s);" % (i,self.ChrLengthMbList)
+ Areas = HT.Area(shape='rect',coords=COORDS,href=HREF)
+ gifmap.areas.append(Areas)
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+ return plotXScale
+
+ # XZ: The only difference of function drawXAxisForPLINK and function drawXAxis are the function name and the self.plotScale condition.
+ def drawXAxisForPLINK(self, fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #Parameters
+ ChrLengthDistList = self.ChrLengthMbList
+ GraphInterval=self.GraphInterval
+
+ NUM_MINOR_TICKS = 5 # Number of minor ticks between major ticks
+ X_MAJOR_TICK_THICKNESS = 2
+ X_MINOR_TICK_THICKNESS = 1
+ X_AXIS_THICKNESS = 1*zoom
+
+ # ======= Alex: Draw the X-axis labels (megabase location)
+ MBLabelFont = pid.Font(ttf="verdana", size=12*fontZoom, bold=0)
+ xMajorTickHeight = 15 # How high the tick extends below the axis
+ xMinorTickHeight = 5*zoom
+ xAxisTickMarkColor = pid.black
+ xAxisLabelColor = pid.black
+ fontHeight = 12*fontZoom # How tall the font that we're using is
+ spacingFromLabelToAxis = 20
+ spacingFromLineToLabel = 3
+
+ if self.plotScale == 'physic':
+ strYLoc = yZero + spacingFromLabelToAxis + canvas.fontHeight(MBLabelFont)
+ ###Physical single chromosome view
+ if self.selectedChr > -1:
+ graphMbWidth = endMb - startMb
+ XScale = Plot.detScale(startMb, endMb)
+ XStart, XEnd, XStep = XScale
+ if XStep < 8:
+ XStep *= 2
+ spacingAmtX = spacingAmt = (XEnd-XStart)/XStep
+
+ j = 0
+ while abs(spacingAmtX -int(spacingAmtX)) >= spacingAmtX/100.0 and j < 6:
+ j += 1
+ spacingAmtX *= 10
+
+ formatStr = '%%2.%df' % j
+
+ for counter, _Mb in enumerate(Plot.frange(XStart, XEnd, spacingAmt / NUM_MINOR_TICKS)):
+ if _Mb < startMb or _Mb > endMb:
+ continue
+ Xc = xLeftOffset + plotXScale*(_Mb - startMb)
+ if counter % NUM_MINOR_TICKS == 0: # Draw a MAJOR mark, not just a minor tick mark
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMajorTickHeight, color=xAxisTickMarkColor, width=X_MAJOR_TICK_THICKNESS) # Draw the MAJOR tick mark
+ labelStr = str(formatStr % _Mb) # What Mbase location to put on the label
+ strWidth = canvas.stringWidth(labelStr, font=MBLabelFont)
+ drawStringXc = (Xc - (strWidth / 2.0))
+ canvas.drawString(labelStr, drawStringXc, strYLoc, font=MBLabelFont, color=xAxisLabelColor, angle=0)
+ else:
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMinorTickHeight, color=xAxisTickMarkColor, width=X_MINOR_TICK_THICKNESS) # Draw the MINOR tick mark
+ # end else
+
+ ###Physical genome wide view
+ else:
+ distScale = 0
+ startPosX = xLeftOffset
+ for i, distLen in enumerate(ChrLengthDistList):
+ if distScale == 0: #universal scale in whole genome mapping
+ if distLen > 75:
+ distScale = 25
+ elif distLen > 30:
+ distScale = 10
+ else:
+ distScale = 5
+ for tickdists in range(distScale, ceil(distLen), distScale):
+ canvas.drawLine(startPosX + tickdists*plotXScale, yZero, startPosX + tickdists*plotXScale, yZero + 7, color=pid.black, width=1*zoom)
+ canvas.drawString(str(tickdists), startPosX+tickdists*plotXScale, yZero + 10*zoom, color=pid.black, font=MBLabelFont, angle=270)
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ megabaseLabelFont = pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ canvas.drawString("Megabases", xLeftOffset + (plotWidth -canvas.stringWidth("Megabases", font=megabaseLabelFont))/2,
+ strYLoc + canvas.fontHeight(MBLabelFont) + 5*zoom, font=megabaseLabelFont, color=pid.black)
+ pass
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset+plotWidth, yZero, color=pid.black, width=X_AXIS_THICKNESS) # Draw the X axis itself
+
+ def drawXAxis(self, fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #Parameters
+ NUM_MINOR_TICKS = 5 # Number of minor ticks between major ticks
+ X_MAJOR_TICK_THICKNESS = 2
+ X_MINOR_TICK_THICKNESS = 1
+ X_AXIS_THICKNESS = 1*zoom
+
+ # ======= Alex: Draw the X-axis labels (megabase location)
+ MBLabelFont = pid.Font(ttf="verdana", size=12*fontZoom, bold=0)
+ xMajorTickHeight = 15 # How high the tick extends below the axis
+ xMinorTickHeight = 5*zoom
+ xAxisTickMarkColor = pid.black
+ xAxisLabelColor = pid.black
+ fontHeight = 12*fontZoom # How tall the font that we're using is
+ spacingFromLabelToAxis = 20
+ spacingFromLineToLabel = 3
+
+ if self.plotScale == 'physic':
+ strYLoc = yZero + spacingFromLabelToAxis + canvas.fontHeight(MBLabelFont)
+ ###Physical single chromosome view
+ if self.selectedChr > -1:
+ graphMbWidth = endMb - startMb
+ XScale = Plot.detScale(startMb, endMb)
+ XStart, XEnd, XStep = XScale
+ if XStep < 8:
+ XStep *= 2
+ spacingAmtX = spacingAmt = (XEnd-XStart)/XStep
+
+ j = 0
+ while abs(spacingAmtX -int(spacingAmtX)) >= spacingAmtX/100.0 and j < 6:
+ j += 1
+ spacingAmtX *= 10
+
+ formatStr = '%%2.%df' % j
+
+ for counter, _Mb in enumerate(Plot.frange(XStart, XEnd, spacingAmt / NUM_MINOR_TICKS)):
+ if _Mb < startMb or _Mb > endMb:
+ continue
+ Xc = xLeftOffset + plotXScale*(_Mb - startMb)
+ if counter % NUM_MINOR_TICKS == 0: # Draw a MAJOR mark, not just a minor tick mark
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMajorTickHeight, color=xAxisTickMarkColor, width=X_MAJOR_TICK_THICKNESS) # Draw the MAJOR tick mark
+ labelStr = str(formatStr % _Mb) # What Mbase location to put on the label
+ strWidth = canvas.stringWidth(labelStr, font=MBLabelFont)
+ drawStringXc = (Xc - (strWidth / 2.0))
+ canvas.drawString(labelStr, drawStringXc, strYLoc, font=MBLabelFont, color=xAxisLabelColor, angle=0)
+ else:
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMinorTickHeight, color=xAxisTickMarkColor, width=X_MINOR_TICK_THICKNESS) # Draw the MINOR tick mark
+ # end else
+
+ ###Physical genome wide view
+ else:
+ distScale = 0
+ startPosX = xLeftOffset
+ for i, distLen in enumerate(self.ChrLengthDistList):
+ if distScale == 0: #universal scale in whole genome mapping
+ if distLen > 75:
+ distScale = 25
+ elif distLen > 30:
+ distScale = 10
+ else:
+ distScale = 5
+ for tickdists in range(distScale, ceil(distLen), distScale):
+ canvas.drawLine(startPosX + tickdists*plotXScale, yZero, startPosX + tickdists*plotXScale, yZero + 7, color=pid.black, width=1*zoom)
+ canvas.drawString(str(tickdists), startPosX+tickdists*plotXScale, yZero + 10*zoom, color=pid.black, font=MBLabelFont, angle=270)
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+ megabaseLabelFont = pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ canvas.drawString("Megabases", xLeftOffset + (plotWidth -canvas.stringWidth("Megabases", font=megabaseLabelFont))/2,
+ strYLoc + canvas.fontHeight(MBLabelFont) + 5*zoom, font=megabaseLabelFont, color=pid.black)
+ pass
+ else:
+ ChrAInfo = []
+ preLpos = -1
+ distinctCount = 0.0
+ if len(self.genotype) > 1:
+ for i, _chr in enumerate(self.genotype):
+ thisChr = []
+ Locus0CM = _chr[0].cM
+ nLoci = len(_chr)
+ if nLoci <= 8:
+ for _locus in _chr:
+ if _locus.name != ' - ':
+ if _locus.cM != preLpos:
+ distinctCount += 1
+ preLpos = _locus.cM
+ thisChr.append([_locus.name, _locus.cM-Locus0CM])
+ else:
+ for j in (0, nLoci/4, nLoci/2, nLoci*3/4, -1):
+ while _chr[j].name == ' - ':
+ j += 1
+ if _chr[j].cM != preLpos:
+ distinctCount += 1
+ preLpos = _chr[j].cM
+ thisChr.append([_chr[j].name, _chr[j].cM-Locus0CM])
+ ChrAInfo.append(thisChr)
+ else:
+ for i, _chr in enumerate(self.genotype):
+ thisChr = []
+ Locus0CM = _chr[0].cM
+ for _locus in _chr:
+ if _locus.name != ' - ':
+ if _locus.cM != preLpos:
+ distinctCount += 1
+ preLpos = _locus.cM
+ thisChr.append([_locus.name, _locus.cM-Locus0CM])
+ ChrAInfo.append(thisChr)
+
+ stepA = (plotWidth+0.0)/distinctCount
+
+ LRectWidth = 10
+ LRectHeight = 3
+ offsetA = -stepA
+ lineColor = pid.lightblue
+ startPosX = xLeftOffset
+ for j, ChrInfo in enumerate(ChrAInfo):
+ preLpos = -1
+ for i, item in enumerate(ChrInfo):
+ Lname,Lpos = item
+ if Lpos != preLpos:
+ offsetA += stepA
+ differ = 1
+ else:
+ differ = 0
+ preLpos = Lpos
+ Lpos *= plotXScale
+ if self.selectedChr > -1:
+ Zorder = i % 5
+ else:
+ Zorder = 0
+ if differ:
+ canvas.drawLine(startPosX+Lpos,yZero,xLeftOffset+offsetA,\
+ yZero+25, color=lineColor)
+ canvas.drawLine(xLeftOffset+offsetA,yZero+25,xLeftOffset+offsetA,\
+ yZero+40+Zorder*(LRectWidth+3),color=lineColor)
+ rectColor = pid.orange
+ else:
+ canvas.drawLine(xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3)-3,\
+ xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3),color=lineColor)
+ rectColor = pid.deeppink
+ canvas.drawRect(xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3),\
+ xLeftOffset+offsetA-LRectHeight,yZero+40+Zorder*(LRectWidth+3)+LRectWidth,\
+ edgeColor=rectColor,fillColor=rectColor,edgeWidth = 0)
+ COORDS="%d,%d,%d,%d"%(xLeftOffset+offsetA-LRectHeight, yZero+40+Zorder*(LRectWidth+3),\
+ xLeftOffset+offsetA,yZero+40+Zorder*(LRectWidth+3)+LRectWidth)
+ HREF="javascript:showDatabase3('%s','%s','%s','');" % (showLocusForm,fd.RISet+"Geno", Lname)
+ Areas=HT.Area(shape='rect',coords=COORDS,href=HREF, title="Locus : " + Lname)
+ gifmap.areas.append(Areas)
+ ##piddle bug
+ if j == 0:
+ canvas.drawLine(startPosX,yZero,startPosX,yZero+40, color=lineColor)
+ startPosX += (self.ChrLengthDistList[j]+self.GraphInterval)*plotXScale
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset+plotWidth, yZero, color=pid.black, width=X_AXIS_THICKNESS) # Draw the X axis itself
+
+ def getColorForMarker(self, chrCount,flag):# no change is needed
+ chrColorDict={}
+ for i in range(chrCount):
+ if flag==1: # display blue and lightblue intercross
+ chrColorDict[i]=pid.black
+ elif flag==0:
+ if (i%2==0):
+ chrColorDict[i]=pid.blue
+ else:
+ chrColorDict[i]=pid.lightblue
+ else:#display different color for different chr
+ if i in [0,8,16]:
+ chrColorDict[i]=pid.black
+ elif i in [1,9,17]:
+ chrColorDict[i]=pid.red
+ elif i in [2,10,18]:
+ chrColorDict[i]=pid.lightgreen
+ elif i in [3,11,19]:
+ chrColorDict[i]=pid.blue
+ elif i in [4,12]:
+ chrColorDict[i]=pid.lightblue
+ elif i in [5,13]:
+ chrColorDict[i]=pid.hotpink
+ elif i in [6,14]:
+ chrColorDict[i]=pid.gold
+ elif i in [7,15]:
+ chrColorDict[i]=pid.grey
+
+ return chrColorDict
+
+
+ def drawProbeSetPosition(self, canvas, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ if len(self.traitList) != 1:
+ return
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ try:
+ Chr = self.traitList[0].chr # self.traitListChr =self.traitList[0].chr=_vals need to change to chrList and mbList
+ Mb = self.traitList[0].mb # self.traitListMb =self.traitList[0].mb=_vals
+ except:
+ return
+
+ if self.plotScale == 'physic':
+ if self.selectedChr > -1:
+ if self.genotype[0].name != Chr or Mb < self.startMb or Mb > self.endMb:
+ return
+ else:
+ locPixel = xLeftOffset + (Mb-self.startMb)*plotXScale
+ else:
+ locPixel = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ if _chr.name != Chr:
+ locPixel += (self.ChrLengthDistList[i] + self.GraphInterval)*plotXScale
+ else:
+ locPixel += Mb*plotXScale
+ break
+ else:
+ if self.selectedChr > -1:
+ if self.genotype[0].name != Chr:
+ return
+ else:
+ for i, _locus in enumerate(self.genotype[0]):
+ #the trait's position is on the left of the first genotype
+ if i==0 and _locus.Mb >= Mb:
+ locPixel=-1
+ break
+
+ #the trait's position is between two traits
+ if i > 0 and self.genotype[0][i-1].Mb < Mb and _locus.Mb >= Mb:
+ locPixel = xLeftOffset + plotXScale*(self.genotype[0][i-1].cM+(_locus.cM-self.genotype[0][i-1].cM)*(Mb -self.genotype[0][i-1].Mb)/(_locus.Mb-self.genotype[0][i-1].Mb))
+ break
+
+ #the trait's position is on the right of the last genotype
+ if i==len(self.genotype[0]) and Mb>=_locus.Mb:
+ locPixel = -1
+ else:
+ locPixel = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ if _chr.name != Chr:
+ locPixel += (self.ChrLengthDistList[i] + self.GraphInterval)*plotXScale
+ else:
+ locPixel += (Mb*(_chr[-1].cM-_chr[0].cM)/self.ChrLengthCMList[i])*plotXScale
+ break
+ if locPixel >= 0:
+ traitPixel = ((locPixel, yZero), (locPixel-6, yZero+12), (locPixel+6, yZero+12))
+ canvas.drawPolygon(traitPixel, edgeColor=pid.black, fillColor=self.TRANSCRIPT_LOCATION_COLOR, closed=1)
+
+ if self.legendChecked:
+ startPosY = 15
+ nCol = 2
+ smallLabelFont = pid.Font(ttf="trebuc", size=12, bold=1)
+ leftOffset = xLeftOffset+(nCol-1)*200
+ canvas.drawPolygon(((leftOffset+6, startPosY-6), (leftOffset, startPosY+6), (leftOffset+12, startPosY+6)), edgeColor=pid.black, fillColor=self.TRANSCRIPT_LOCATION_COLOR, closed=1)
+ canvas.drawString("Sequence Site", (leftOffset+15), (startPosY+5), smallLabelFont, self.TOP_RIGHT_INFO_COLOR)
+
+ # build dict based on plink result, key is chr, value is list of [snp,BP,pValue]
+ def getPlinkResultDict(self,outputFileName='',thresholdPvalue=-1,ChrOrderIdNameDict={}):
+
+ ChrList =self.ChrList
+ plinkResultDict={}
+
+ plinkResultfp = open("%s%s.qassoc"% (webqtlConfig.TMPDIR, outputFileName), "rb")
+
+ headerLine=plinkResultfp.readline()# read header line
+ line = plinkResultfp.readline()
+
+ valueList=[] # initialize value list, this list will include snp, bp and pvalue info
+ pValueList=[]
+ count=0
+
+ while line:
+ #convert line from str to list
+ lineList=self.buildLineList(line=line)
+
+ # only keep the records whose chromosome name is in db
+ if ChrOrderIdNameDict.has_key(int(lineList[0])) and lineList[-1] and lineList[-1].strip()!='NA':
+
+ chrName=ChrOrderIdNameDict[int(lineList[0])]
+ snp = lineList[1]
+ BP = lineList[2]
+ pValue = float(lineList[-1])
+ pValueList.append(pValue)
+
+ if plinkResultDict.has_key(chrName):
+ valueList=plinkResultDict[chrName]
+
+ # pvalue range is [0,1]
+ if thresholdPvalue >=0 and thresholdPvalue<=1:
+ if pValue < thresholdPvalue:
+ valueList.append((snp,BP,pValue))
+ count+=1
+
+ plinkResultDict[chrName]=valueList
+ valueList=[]
+ else:
+ if thresholdPvalue>=0 and thresholdPvalue<=1:
+ if pValue < thresholdPvalue:
+ valueList.append((snp,BP,pValue))
+ count+=1
+
+ if valueList:
+ plinkResultDict[chrName]=valueList
+
+ valueList=[]
+
+
+ line =plinkResultfp.readline()
+ else:
+ line=plinkResultfp.readline()
+
+ if pValueList:
+ minPvalue= min(pValueList)
+ else:
+ minPvalue=0
+
+ return count,minPvalue,plinkResultDict
+
+
+ ######################################################
+ # input: line: str,one line read from file
+ # function: convert line from str to list;
+ # output: lineList list
+ #######################################################
+ def buildLineList(self,line=None):
+
+ lineList = string.split(string.strip(line),' ')# irregular number of whitespaces between columns
+ lineList =[ item for item in lineList if item <>'']
+ lineList = map(string.strip, lineList)
+
+ return lineList
+
+ #added by NL: automatically generate pheno txt file for PLINK based on strainList passed from dataEditing page
+ def genPhenoTxtFileForPlink(self,phenoFileName='', RISetName='', probesetName='', valueDict={}):
+ pedFileStrainList=self.getStrainNameFromPedFile(RISetName=RISetName)
+ outputFile = open("%s%s.txt"%(webqtlConfig.TMPDIR,phenoFileName),"wb")
+ headerLine = 'FID\tIID\t%s\n'%probesetName
+ outputFile.write(headerLine)
+
+ newValueList=[]
+
+ #if valueDict does not include some strain, value will be set to -9999 as missing value
+ for item in pedFileStrainList:
+ try:
+ value=valueDict[item]
+ value=str(value).replace('value=','')
+ value=value.strip()
+ except:
+ value=-9999
+
+ newValueList.append(value)
+
+
+ newLine=''
+ for i, strain in enumerate(pedFileStrainList):
+ j=i+1
+ value=newValueList[i]
+ newLine+='%s\t%s\t%s\n'%(strain, strain, value)
+
+ if j%1000==0:
+ outputFile.write(newLine)
+ newLine=''
+
+ if newLine:
+ outputFile.write(newLine)
+
+ outputFile.close()
+
+ # get strain name from ped file in order
+ def getStrainNameFromPedFile(self, RISetName=''):
+ pedFileopen= open("%splink/%s.ped"%(webqtlConfig.HTMLPATH, RISetName),"r")
+ line =pedFileopen.readline()
+ strainNameList=[]
+
+ while line:
+ lineList=string.split(string.strip(line),'\t')
+ lineList=map(string.strip,lineList)
+
+ strainName=lineList[0]
+ strainNameList.append(strainName)
+
+ line =pedFileopen.readline()
+
+ return strainNameList
+
+ ################################################################
+ # Generate Chr list, Chr OrderId and Retrieve Length Information
+ ################################################################
+ def getChrNameOrderIdLength(self,RISet=''):
+
+ try:
+ query = """
+ Select
+ Chr_Length.Name,Chr_Length.OrderId,Length from Chr_Length, InbredSet
+ where
+ Chr_Length.SpeciesId = InbredSet.SpeciesId AND
+ InbredSet.Name = '%s'
+ Order by OrderId
+ """ % (RISet)
+ self.cursor.execute(query)
+
+ results =self.cursor.fetchall()
+ ChrList=[]
+ ChrLengthMbList=[]
+ ChrNameOrderIdDict={}
+ ChrOrderIdNameDict={}
+
+ for item in results:
+ ChrList.append(item[0])
+ ChrNameOrderIdDict[item[0]]=item[1] # key is chr name, value is orderId
+ ChrOrderIdNameDict[item[1]]=item[0] # key is orderId, value is chr name
+ ChrLengthMbList.append(item[2])
+
+ except:
+ ChrList=[]
+ ChrNameOrderIdDict={}
+ ChrLengthMbList=[]
+
+ return ChrList,ChrNameOrderIdDict,ChrOrderIdNameDict,ChrLengthMbList
diff --git a/wqflask/wqflask/marker_regression/__init__.py b/wqflask/wqflask/marker_regression/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/wqflask/wqflask/marker_regression/marker_regression.py b/wqflask/wqflask/marker_regression/marker_regression.py
new file mode 100755
index 00000000..ed01a3fa
--- /dev/null
+++ b/wqflask/wqflask/marker_regression/marker_regression.py
@@ -0,0 +1,1648 @@
+from __future__ import absolute_import, print_function, division
+
+from base.trait import GeneralTrait
+from base import data_set #import create_dataset
+
+from pprint import pformat as pf
+
+import time
+import string
+import math
+#from math import *
+#import piddle
+import sys
+import os
+import httplib
+import urllib
+
+from htmlgen import HTMLgen2 as HT
+from utility import Plot
+from wqflask.interval_analyst import GeneUtil
+from base.trait import GeneralTrait
+from base.data_set import create_dataset
+from base.templatePage import templatePage
+from utility import webqtlUtil
+from base import webqtlConfig
+from dbFunction import webqtlDatabaseFunction
+from base.GeneralObject import GeneralObject
+
+import reaper
+import cPickle
+from utility.THCell import THCell
+from utility.TDCell import TDCell
+
+
+class MarkerRegression(object):
+
+ #def __init__(self, start_vars):
+ #
+ # print("[mike] Now start_vars is:", pf(start_vars))
+ #
+ # self.dataset = data_set.create_dataset(start_vars['dataset_name'])
+ # self.this_trait = GeneralTrait(dataset=self.dataset.name,
+ # name=start_vars['trait_id'],
+ # cellid=None)
+ #
+ # print("self.this_trait is: ", pf(self.this_trait))
+ # print("self.dataset is: ", pf(self.dataset))
+
+ def __init__(self, start_vars):
+ #templatePage.__init__(self, fd)
+
+ #if not self.openMysql():
+ # return
+
+ self.dataset = create_dataset(start_vars['dataset_name'])
+
+ #self.initializeParameters(start_vars)
+
+ #filename= webqtlUtil.genRandStr("Itvl_")
+ #ChrList,ChrNameOrderIdDict,ChrOrderIdNameDict,ChrtLengthMbList= self.getChrNameOrderIdLength(RISet=fd.RISet)
+
+ if False: # For PLINK
+
+ traitInfoList = string.split(string.strip(fd.identification),':')
+ probesetName = string.strip(traitInfoList[-1])
+ plinkOutputFileName= webqtlUtil.genRandStr("%s_%s_"%(fd.RISet,probesetName))
+
+ # get related values from fd.allTraitData; the format of 'allTraitValueDict'is {strainName1: value=-0.2...}
+ fd.readData()
+ allTraitValueDict = fd.allTraitData
+
+ #automatically generate pheno txt file for PLINK
+ self.genPhenoTxtFileForPlink(phenoFileName=plinkOutputFileName,RISetName=fd.RISet,probesetName=probesetName, valueDict=allTraitValueDict)
+ # os.system full path is required for input and output files; specify missing value is -9999
+ plink_command = '%splink/plink --noweb --ped %splink/%s.ped --no-fid --no-parents --no-sex --no-pheno --map %splink/%s.map --pheno %s/%s.txt --pheno-name %s --missing-phenotype -9999 --out %s%s --assoc ' % (webqtlConfig.HTMLPATH, webqtlConfig.HTMLPATH, fd.RISet, webqtlConfig.HTMLPATH, fd.RISet, webqtlConfig.TMPDIR, plinkOutputFileName, probesetName, webqtlConfig.TMPDIR, plinkOutputFileName)
+
+ os.system(plink_command)
+
+ if fd.identification:
+ heading2 = HT.Paragraph('Trait ID: %s' % fd.identification)
+ heading2.__setattr__("class","subtitle")
+ self.dict['title'] = '%s: Genome Association' % fd.identification
+ else:
+ heading2 = ""
+ self.dict['title'] = 'Genome Association'
+
+ if fd.traitInfo:
+ symbol,chromosome,MB = string.split(fd.traitInfo,'\t')
+ heading3 = HT.Paragraph('[ ',HT.Strong(HT.Italic('%s' % symbol,id="green")),' on Chr %s @ %s Mb ]' % (chromosome,MB))
+ else:
+ heading3 = ""
+
+ heading = HT.Paragraph('Trait Data Entered for %s Set' % fd.RISet)
+ heading.__setattr__("class","title")
+
+ # header info part:Trait Data Entered for HLC Set & Trait ID:
+ headerdiv = HT.TR(HT.TD(heading, heading2,heading3, width='45%',valign='top', align='left', bgColor='#eeeeee'))
+
+ self.ChrList=ChrList # get chr name from '1' to 'X'
+ self.ChrLengthMbList = ChrLengthMbList
+
+ # build plink result dict based on chr, key is chr name, value is in list type including Snpname, bp and pvalue info
+ plinkResultDict={}
+ count,minPvalue,plinkResultDict =self.getPlinkResultDict(outputFileName=plinkOutputFileName,thresholdPvalue=self.pValue,ChrOrderIdNameDict=ChrOrderIdNameDict)
+
+ # if can not find results which are matched with assigned p-value, system info will show up
+ if count >0:
+
+ #for genome association report table
+ reportTable=""
+ # sortable table object
+ resultstable,tblobj,bottomInfo = self.GenReportForPLINK(ChrNameOrderIdDict=ChrNameOrderIdDict, RISet=fd.RISet,plinkResultDict=plinkResultDict,thresholdPvalue=self.pValue,chrList=self.ChrList)
+
+ # creat object for result table for sort function
+ objfile = open('%s.obj' % (webqtlConfig.TMPDIR+filename), 'wb')
+ cPickle.dump(tblobj, objfile)
+ objfile.close()
+
+ sortby = ("Index", "up")
+ reportTable =HT.Div(webqtlUtil.genTableObj(tblobj=tblobj, file=filename, sortby=sortby, tableID = "sortable", addIndex = "0"), Id="sortable")
+
+ descriptionTable = HT.TableLite(border=0, cellpadding=0, cellspacing=0)
+ descriptionTable.append(HT.TR(HT.TD(reportTable, colspan=3)))
+ descriptionTable.append(HT.TR(HT.TD(HT.BR(),HT.BR())))
+ descriptionTable.append(bottomInfo)
+
+ # get each chr's length
+ self.ChrLengthMbList = map(lambda x: x/1000000.0, self.ChrLengthMbList) # change unit from bp to mb
+ self.ChrLengthMbSum = reduce(lambda x, y:x+y, self.ChrLengthMbList, 0.0)# get total length of all chrs
+ if self.ChrLengthMbList:
+ self.GraphInterval = self.ChrLengthMbSum/(len(self.ChrLengthMbList)*12) #Empirical Mb interval
+ else:
+ self.GraphInterval = 1
+
+ # for human data, there's no CM value
+ self.ChrLengthCMList = []
+ self.ChrLengthCMSum = 0
+
+ # begin: common part with human data
+ intCanvas = pid.PILCanvas(size=(self.graphWidth,self.graphHeight))
+ gifmap = self.plotIntMappingForPLINK(fd, intCanvas, startMb = self.startMb, endMb = self.endMb, plinkResultDict=plinkResultDict)
+
+ intCanvas.save(os.path.join(webqtlConfig.IMGDIR, filename), format='png')
+ intImg=HT.Image('/image/'+filename+'.png', border=0, usemap='#WebQTLImageMap')
+
+ TD_LR = HT.TR(HT.TD(HT.Blockquote(gifmap,intImg, HT.P()), bgColor='#eeeeee', height = 200))
+ self.dict['body'] = str(headerdiv)+str(TD_LR)+str(resultstable)+str(HT.TR(HT.TD(descriptionTable)))
+
+ else:
+ heading = "Genome Association"
+ detail = ['There is no association with marker that meets this criteria. Please provide a less stringend threshold. The minimun p-value is %s.'%minPvalue]
+ self.error(heading=heading,detail=detail)
+ return
+
+ else: # QTLreaper result
+ #if not fd.genotype:
+ # fd.readData()
+ #
+ #fd.parentsf14regression = fd.formdata.getvalue('parentsf14regression')
+ #weightedRegression = fd.formdata.getvalue('applyVarianceSE')
+
+ #if fd.parentsf14regression and fd.genotype_2:
+ # _genotype = fd.genotype_2
+ #else:
+ genotype = self.dataset.group.read_genotype_file()
+ print("[black]:", genotype)
+
+ _strains, _vals, _vars, N = fd.informativeStrains(_genotype.prgy, weightedRegression)
+
+ if fd.identification:
+ heading2 = HT.Paragraph('Trait ID: %s' % fd.identification)
+ heading2.__setattr__("class","subtitle")
+ self.dict['title'] = '%s: Genome Association' % fd.identification
+ else:
+ heading2 = ""
+ self.dict['title'] = 'Genome Association'
+
+ if fd.traitInfo:
+ symbol,chromosome,MB = string.split(fd.traitInfo,'\t')
+ heading3 = HT.Paragraph('[ ',HT.Strong(HT.Italic('%s' % symbol,id="green")),' on Chr %s @ %s Mb ]' % (chromosome,MB))
+ else:
+ heading3 = ""
+
+ if N < webqtlConfig.KMININFORMATIVE:
+ heading = "Genome Association"
+ detail = ['Fewer than %d strain data were entered for %s data set. No mapping attempted.' % (webqtlConfig.KMININFORMATIVE, fd.RISet)]
+ self.error(heading=heading,detail=detail)
+ return
+ else:
+ heading = HT.Paragraph('Trait Data Entered for %s Set' % fd.RISet)
+ heading.__setattr__("class","title")
+
+ datadiv = HT.TD(heading, heading2,heading3, width='45%',valign='top', align='left', bgColor='#eeeeee')
+ resultstable,tblobj,bottomInfo = self.GenReport(ChrNameOrderIdDict,fd, _genotype, _strains, _vals, _vars)
+ #resultstable = self.GenReport(fd, _genotype, _strains, _vals, _vars)
+
+ # creat object for result table for sort function
+ objfile = open('%s.obj' % (webqtlConfig.TMPDIR+filename), 'wb')
+ cPickle.dump(tblobj, objfile)
+ objfile.close()
+
+ sortby = ("Index", "up")
+ reportTable =HT.Div(webqtlUtil.genTableObj(tblobj=tblobj, file=filename, sortby=sortby, tableID = "sortable", addIndex = "0"), Id="sortable")
+
+ descriptionTable = HT.TableLite(border=0, cellpadding=0, cellspacing=0)
+ descriptionTable.append(HT.TR(HT.TD(reportTable, colspan=3)))
+ descriptionTable.append(HT.TR(HT.TD(HT.BR(),HT.BR())))
+ descriptionTable.append(bottomInfo)
+
+ self.traitList=_vals
+
+ ##########################plot#######################
+
+ ################################################################
+ # Generate Chr list and Retrieve Length Information
+ ################################################################
+ self.genotype= _genotype
+ self.ChrList = [("All", -1)]
+
+ for i, indChr in enumerate(self.genotype):
+ self.ChrList.append((indChr.name, i))
+
+ self.cursor.execute("""
+ Select
+ Length from Chr_Length, InbredSet
+ where
+ Chr_Length.SpeciesId = InbredSet.SpeciesId AND
+ InbredSet.Name = '%s' AND
+ Chr_Length.Name in (%s)
+ Order by
+ OrderId
+ """ % (fd.RISet, string.join(map(lambda X: "'%s'" % X[0], self.ChrList[1:]), ", ")))
+
+ self.ChrLengthMbList = self.cursor.fetchall()
+ self.ChrLengthMbList = map(lambda x: x[0]/1000000.0, self.ChrLengthMbList)
+ self.ChrLengthMbSum = reduce(lambda x, y:x+y, self.ChrLengthMbList, 0.0)
+ if self.ChrLengthMbList:
+ self.MbGraphInterval = self.ChrLengthMbSum/(len(self.ChrLengthMbList)*12) #Empirical Mb interval
+ else:
+ self.MbGraphInterval = 1
+
+ self.ChrLengthCMList = []
+ for i, _chr in enumerate(self.genotype):
+ self.ChrLengthCMList.append(_chr[-1].cM - _chr[0].cM)
+ self.ChrLengthCMSum = reduce(lambda x, y:x+y, self.ChrLengthCMList, 0.0)# used for calculate plot scale
+
+ self.GraphInterval = self.MbGraphInterval #Mb
+
+ # begin: common part with human data
+ intCanvas = pid.PILCanvas(size=(self.graphWidth,self.graphHeight))
+ gifmap = self.plotIntMapping(fd, intCanvas, startMb = self.startMb, endMb = self.endMb, showLocusForm= "")
+ filename= webqtlUtil.genRandStr("Itvl_")
+ intCanvas.save(os.path.join(webqtlConfig.IMGDIR, filename), format='png')
+ intImg=HT.Image('/image/'+filename+'.png', border=0, usemap='#WebQTLImageMap')
+
+ ################################################################
+ # footnote goes here
+ ################################################################
+ btminfo = HT.Paragraph(Id="smallsize") #Small('More information about this graph is available here.')
+
+ if (self.additiveChecked):
+ btminfo.append(HT.BR(), 'A positive additive coefficient (', HT.Font('green', color='green'), ' line) indicates that %s alleles increase trait values. In contrast, a negative additive coefficient (' % fd.ppolar, HT.Font('red', color='red'), ' line) indicates that %s alleles increase trait values.' % fd.mpolar)
+
+
+ TD_LR = HT.TR(HT.TD(HT.Blockquote(gifmap,intImg, HT.P()), bgColor='#eeeeee', height = 200))
+
+ self.dict['body'] = str(datadiv)+str(TD_LR)+str(resultstable)+str(HT.TR(HT.TD(descriptionTable)))
+
+ # end: common part with human data
+
+
+
+
+ # add by NL 10-2-2011
+ def initializeParameters(self, fd):
+ """
+ Initializes all of the MarkerRegressionPage class parameters,
+ acquiring most values from the formdata (fd)
+ """
+ ###################################
+ # manhattam plot parameters
+ ###################################
+
+ self.graphHeight = 600
+ self.graphWidth = 1280
+ self.plotScale = 'physic'
+ self.selectedChr = -1
+ self.GRAPH_BACK_DARK_COLOR = pid.HexColor(0xF1F1F9)
+ self.GRAPH_BACK_LIGHT_COLOR = pid.HexColor(0xFBFBFF)
+ self.LRS_COLOR = pid.HexColor(0x0000FF)
+ self.LRS_LOD ='LRS'
+ self.lrsMax = float(fd.formdata.getvalue('lrsMax', 0))
+ self.startMb = fd.formdata.getvalue('startMb', "-1")
+ self.endMb = fd.formdata.getvalue('endMb', "-1")
+ self.mappingMethodId = fd.formdata.getvalue('mappingMethodId', "0")
+ self.permChecked=True
+ self.multipleInterval=False
+ self.SIGNIFICANT_WIDTH = 5
+ self.SUGGESTIVE_WIDTH = 5
+ self.SIGNIFICANT_COLOR = pid.HexColor(0xEBC7C7)
+ self.SUGGESTIVE_COLOR = pid.gainsboro
+ self.colorCollection = [self.LRS_COLOR]
+ self.additiveChecked= True
+ self.ADDITIVE_COLOR_POSITIVE = pid.green
+ self.legendChecked =False
+ self.pValue=float(fd.formdata.getvalue('pValue',-1))
+
+ # allow user to input p-value greater than 1,
+ # in this case, the value will be treated as -lgP value. so the input value needs to be transferred to power of 10 format
+ if self.pValue >1:
+ self.pValue =10**-(self.pValue)
+
+ try:
+ self.startMb = float(self.startMb)
+ self.endMb = float(self.endMb)
+ if self.startMb > self.endMb:
+ temp = self.startMb
+ self.startMb = self.endMb
+ self.endMb = temp
+ #minimal distance 10bp
+ if self.endMb - self.startMb < 0.00001:
+ self.endMb = self.startMb + 0.00001
+ except:
+ self.startMb = self.endMb = -1
+
+ def GenReportForPLINK(self, ChrNameOrderIdDict={},RISet='',plinkResultDict= {},thresholdPvalue=-1,chrList=[]):
+
+ 'Create an HTML division which reports any loci which are significantly associated with the submitted trait data.'
+ #########################################
+ # Genome Association report
+ #########################################
+ locusFormName = webqtlUtil.genRandStr("fm_")
+ locusForm = HT.Form(cgi = os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), \
+ enctype='multipart/form-data', name=locusFormName, submit=HT.Input(type='hidden'))
+ hddn = {'FormID':'showDatabase','ProbeSetID':'_','database':RISet+"Geno",'CellID':'_', \
+ 'RISet':RISet, 'incparentsf1':'on'}
+ for key in hddn.keys():
+ locusForm.append(HT.Input(name=key, value=hddn[key], type='hidden'))
+
+ regressionHeading = HT.Paragraph('Genome Association Report')
+ regressionHeading.__setattr__("class","title")
+
+ filename= webqtlUtil.genRandStr("GenomeAsscociation_")
+ fpText = open('%s.txt' % (webqtlConfig.TMPDIR+filename), 'wb')
+ fpText.write('The loci meet the criteria of P-Value <= %3.6f.\n'%thresholdPvalue)
+ pValueInfo =HT.Paragraph('The loci meet the criteria of P-Value <= %3.6f.\n'%thresholdPvalue)
+
+ textUrl = HT.Href(text = 'Download', url= '/tmp/'+filename+'.txt', target = "_blank", Class='fs12 fwn')
+ bottomInfo = HT.TR(HT.TD(HT.Paragraph(textUrl, ' result in tab-delimited text format.', HT.BR(), HT.BR(),Class="fs12 fwn"), colspan=3))
+
+ tblobj={} # build dict for genTableObj function; keys include header and body
+ tblobj_header = [] # value of key 'header'
+ tblobj_body=[] # value of key 'body'
+ reportHeaderRow=[] # header row list for tblobj_header (html part)
+ headerList=['Index','SNP Name','Chr','Mb','-log(P)']
+ headerStyle="fs14 fwb ffl b1 cw cbrb" # style of the header
+ cellColorStyle = "fs13 b1 fwn c222" # style of the cells
+
+ if headerList:
+ for ncol, item in enumerate(headerList):
+ reportHeaderRow.append(THCell(HT.TD(item, Class=headerStyle, valign='bottom',nowrap='ON'),text=item, idx=ncol))
+ #download file for table headers' names
+ fpText.write('SNP_Name\tChromosome\tMb\t-log(P)\n')
+
+ tblobj_header.append(reportHeaderRow)
+ tblobj['header']=tblobj_header
+
+ index=1
+ for chr in chrList:
+
+ if plinkResultDict.has_key(chr):
+ if chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[chr]
+ else:
+ chrOrderId=chr
+
+ valueList=plinkResultDict[chr]
+
+ for value in valueList:
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ snpName=value[0]
+ bp=value[1]
+ mb=int(bp)/1000000.0
+
+ try:
+ pValue =float(value[2])
+ except:
+ pValue =1
+ formattedPvalue = -math.log10(pValue)
+
+ formattedPvalue = webqtlUtil.SciFloat(formattedPvalue)
+ dbSnprs=snpName.replace('rs','')
+ SnpHref = HT.Href(text=snpName, url="http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=%s"%dbSnprs, target="_blank")
+
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(HT.TD(SnpHref, Class=cellColorStyle,nowrap='ON'),snpName, snpName))
+ reportBodyRow.append(TDCell(HT.TD(chr, Class=cellColorStyle, align="center",nowrap='ON'),chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%mb, Class=cellColorStyle, align="center",nowrap='ON'),mb, mb))
+ reportBodyRow.append(TDCell(HT.TD(formattedPvalue, Class=cellColorStyle, align="center",nowrap='ON'),formattedPvalue, float(formattedPvalue)))
+
+ fpText.write('%s\t%s\t%3.6f\t%s\n' % (snpName, str(chr), mb, formattedPvalue))
+ index+=1
+
+ tblobj_body.append(reportBodyRow)
+
+ tblobj['body']=tblobj_body
+ rv=HT.TR(HT.TD(regressionHeading,pValueInfo, locusForm, HT.P(), width='55%',valign='top', align='left',bgColor='#eeeeee'))
+
+ return rv, tblobj,bottomInfo
+
+
+ def GenReport(self, ChrNameOrderIdDict,fd, _genotype, _strains, _vals, _vars= []):
+ 'Create an HTML division which reports any loci which are significantly associated with the submitted trait data.'
+ #calculate QTL for each trait
+ self.qtlresults = []
+ if webqtlUtil.ListNotNull(_vars):
+ qtlresults = _genotype.regression(strains = _strains, trait = _vals, variance = _vars)
+ LRSArray = _genotype.permutation(strains = _strains, trait = _vals, variance = _vars, nperm=fd.nperm)
+ else:
+ qtlresults = _genotype.regression(strains = _strains, trait = _vals)
+ LRSArray = _genotype.permutation(strains = _strains, trait = _vals,nperm=fd.nperm)
+
+ self.qtlresults.append(qtlresults)
+
+ filename= webqtlUtil.genRandStr("GenomeAsscociation_")
+
+ # set suggestive, significant and highly significant LRS
+ if fd.suggestive == None:
+ fd.suggestive = LRSArray[int(fd.nperm*0.37-1)]
+ else:
+ fd.suggestive = float(fd.suggestive)
+ if fd.significance == None:
+ fd.significance = LRSArray[int(fd.nperm*0.95-1)]
+ else:
+ fd.significance = float(fd.significance)
+
+ self.significance =fd.significance
+ self.suggestive = fd.suggestive
+ self.highlysignificant = LRSArray[int(fd.nperm*0.99-1)]
+ _dispAllLRS = 0
+ if fd.formdata.getvalue('displayAllLRS'):
+ _dispAllLRS = 1
+ qtlresults2 = []
+ if _dispAllLRS:
+ filtered = qtlresults[:]
+ else:
+ filtered = filter(lambda x, y=fd.suggestive: x.lrs > y, qtlresults)
+ if len(filtered) == 0:
+ qtlresults2 = qtlresults[:]
+ qtlresults2.sort()
+ filtered = qtlresults2[-10:]
+
+ #########################################
+ # Permutation Graph
+ #########################################
+ myCanvas = pid.PILCanvas(size=(400,300))
+ #plotBar(myCanvas,10,10,390,290,LRSArray,XLabel='LRS',YLabel='Frequency',title=' Histogram of Permutation Test',identification=fd.identification)
+ Plot.plotBar(myCanvas, LRSArray, XLabel='LRS',YLabel='Frequency',title=' Histogram of Permutation Test')
+ filename= webqtlUtil.genRandStr("Reg_")
+ myCanvas.save(webqtlConfig.IMGDIR+filename, format='gif')
+ img=HT.Image('/image/'+filename+'.gif',border=0,alt='Histogram of Permutation Test')
+
+ if fd.suggestive == None:
+ fd.suggestive = LRSArray[int(fd.nperm*0.37-1)]
+ else:
+ fd.suggestive = float(fd.suggestive)
+ if fd.significance == None:
+ fd.significance = LRSArray[int(fd.nperm*0.95-1)]
+ else:
+ fd.significance = float(fd.significance)
+
+ permutationHeading = HT.Paragraph('Histogram of Permutation Test')
+ permutationHeading.__setattr__("class","title")
+
+ permutation = HT.TableLite()
+ permutation.append(HT.TR(HT.TD(img)))
+
+
+ #########################################
+ # Genome Association report
+ #########################################
+ locusFormName = webqtlUtil.genRandStr("fm_")
+ locusForm = HT.Form(cgi = os.path.join(webqtlConfig.CGIDIR, webqtlConfig.SCRIPTFILE), \
+ enctype='multipart/form-data', name=locusFormName, submit=HT.Input(type='hidden'))
+ hddn = {'FormID':'showDatabase','ProbeSetID':'_','database':fd.RISet+"Geno",'CellID':'_', \
+ 'RISet':fd.RISet, 'incparentsf1':'on'}
+ for key in hddn.keys():
+ locusForm.append(HT.Input(name=key, value=hddn[key], type='hidden'))
+
+ regressionHeading = HT.Paragraph('Genome Association Report')
+ regressionHeading.__setattr__("class","title")
+ # report is the info part above report table
+ if qtlresults2 != []:
+ report = HT.Blockquote(HT.Font('No association ',color="#FF0000"),HT.Font('with a likelihood ratio statistic greater than %3.1f was found. Here are the top 10 LRSs.' % fd.suggestive,color="#000000"))
+ else:
+ report = HT.Blockquote('The following loci in the %s data set have associations with the above trait data.\n' % fd.RISet, HT.P())
+ report.__setattr__("class","normalsize")
+
+ fpText = open('%s.txt' % (webqtlConfig.TMPDIR+filename), 'wb')
+ fpText.write('Suggestive LRS =%3.2f\n'%self.suggestive)
+ fpText.write('Significant LRS =%3.2f\n'%self.significance)
+ fpText.write('Highly Significant LRS =%3.2f\n'%self.highlysignificant)
+ LRSInfo =HT.Paragraph(' Suggestive LRS =%3.2f\n'%fd.suggestive, HT.BR(), ' Significant LRS =%3.2f\n'%fd.significance,HT.BR(),' Highly Significant LRS =%3.2f\n' % self.highlysignificant)
+
+ textUrl = HT.Href(text = 'Download', url= '/tmp/'+filename+'.txt', target = "_blank", Class='fs12 fwn')
+
+ bottomInfo = HT.TR(HT.TD(HT.Paragraph(textUrl, ' result in tab-delimited text format.', HT.BR(), HT.BR(),'LRS values marked with',HT.Font(' * ',color="red"), 'are greater than the significance threshold (specified by you or by permutation test). ' , HT.BR(), HT.BR(), HT.Strong('Additive Effect'), ' is half the difference in the mean phenotype of all cases that are homozygous for one parental allel at this marker minus the mean of all cases that are homozygous for the other parental allele at this marker. ','In the case of %s strains, for example,' % fd.RISet,' A positive additive effect indicates that %s alleles increase trait values. Negative additive effect indicates that %s alleles increase trait values.'% (fd.ppolar,fd.mpolar),Class="fs12 fwn")))
+
+ tblobj={} # build dict for genTableObj function; keys include header and body
+ tblobj_header = [] # value of key 'header'
+ tblobj_body=[] # value of key 'body'
+ reportHeaderRow=[] # header row list for tblobj_header (html part)
+ headerStyle="fs14 fwb ffl b1 cw cbrb" # style of the header
+ cellColorStyle = "fs13 b1 fwn c222" # style of the cells
+
+ headerList=['Index','LRS','Chr','Mb','Locus','Additive Effect']
+ for ncol, item in enumerate(headerList):
+ reportHeaderRow.append(THCell(HT.TD(item, Class=headerStyle, valign='bottom',nowrap='ON'),text=item, idx=ncol))
+
+ if fd.genotype.type == 'intercross':
+ ncol =len(headerList)
+ reportHeaderRow.append(THCell(HT.TD('Dominance Effect', Class=headerStyle, valign='bottom',nowrap='ON'),text='Dominance Effect', idx=ncol))
+
+ #download file for table headers' names
+ fpText.write('LRS\tChromosome\tMb\tLocus\tAdditive Effect\tDominance Effect\n')
+
+ index=1
+ for ii in filtered:
+ #add by NL 06-20-2011: set LRS to 460 when LRS is infinite,
+ if ii.lrs==float('inf') or ii.lrs>webqtlConfig.MAXLRS:
+ LRS=webqtlConfig.MAXLRS #maximum LRS value
+ else:
+ LRS=ii.lrs
+
+ if LRS > fd.significance:
+ lrs = HT.TD(HT.Font('%3.3f*' % LRS, color='#FF0000'),Class=cellColorStyle)
+ else:
+ lrs = HT.TD('%3.3f' % LRS,Class=cellColorStyle)
+
+ if ii.locus.chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[ii.locus.chr]
+ else:
+ chrOrderId=ii.locus.chr
+
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(lrs,LRS, LRS))
+ reportBodyRow.append(TDCell(HT.TD(ii.locus.chr, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%ii.locus.Mb, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.Mb, ii.locus.Mb))
+ reportBodyRow.append(TDCell(HT.TD(HT.Href(text=ii.locus.name, url = "javascript:showTrait('%s','%s');" % (locusFormName, ii.locus.name), Class='normalsize'), Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.name, ii.locus.name))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.additive, Class=cellColorStyle, align="center",nowrap='ON'),ii.additive, ii.additive))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.dominance, Class=cellColorStyle, align="center",nowrap='ON'),ii.dominance, ii.dominance))
+
+ fpText.write('%2.3f\t%s\t%3.6f\t%s\t%2.3f\t%2.3f\n' % (LRS, ii.locus.chr, ii.locus.Mb, ii.locus.name, ii.additive, ii.dominance))
+ index+=1
+ tblobj_body.append(reportBodyRow)
+ else:
+ #download file for table headers' names
+ fpText.write('LRS\tChromosome\tMb\tLocus\tAdditive Effect\n')
+
+ index=1
+ for ii in filtered:
+ #add by NL 06-20-2011: set LRS to 460 when LRS is infinite,
+ if ii.lrs==float('inf') or ii.lrs>webqtlConfig.MAXLRS:
+ LRS=webqtlConfig.MAXLRS #maximum LRS value
+ else:
+ LRS=ii.lrs
+
+ if LRS > fd.significance:
+ lrs = HT.TD(HT.Font('%3.3f*' % LRS, color='#FF0000'),Class=cellColorStyle)
+ else:
+ lrs = HT.TD('%3.3f' % LRS,Class=cellColorStyle)
+
+ if ii.locus.chr in ChrNameOrderIdDict.keys():
+ chrOrderId =ChrNameOrderIdDict[ii.locus.chr]
+ else:
+ chrOrderId=ii.locus.chr
+
+ reportBodyRow=[] # row list for tblobj_body (html part)
+ selectCheck=HT.Input(type="checkbox", Class="checkbox", name="index",value=index, onClick="highlight(this)")
+ reportBodyRow.append(TDCell(HT.TD(str(index),selectCheck, align='right',Class=cellColorStyle,nowrap='ON'),str(index),index))
+ reportBodyRow.append(TDCell(lrs,LRS, LRS))
+ reportBodyRow.append(TDCell(HT.TD(ii.locus.chr, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.chr, chrOrderId))
+ reportBodyRow.append(TDCell(HT.TD('%3.6f'%ii.locus.Mb, Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.Mb, ii.locus.Mb))
+ reportBodyRow.append(TDCell(HT.TD(HT.Href(text=ii.locus.name, url = "javascript:showTrait('%s','%s');" % (locusFormName, ii.locus.name), Class='normalsize'), Class=cellColorStyle, align="center",nowrap='ON'),ii.locus.name, ii.locus.name))
+ reportBodyRow.append(TDCell(HT.TD('%3.3f' % ii.additive, Class=cellColorStyle, align="center",nowrap='ON'),ii.additive, ii.additive))
+
+ fpText.write('%2.3f\t%s\t%3.6f\t%s\t%2.3f\n' % (LRS, ii.locus.chr, ii.locus.Mb, ii.locus.name, ii.additive))
+ index+=1
+ tblobj_body.append(reportBodyRow)
+
+ tblobj_header.append(reportHeaderRow)
+ tblobj['header']=tblobj_header
+ tblobj['body']=tblobj_body
+
+ rv=HT.TD(regressionHeading,LRSInfo,report, locusForm, HT.P(),width='55%',valign='top', align='left', bgColor='#eeeeee')
+ if fd.genotype.type == 'intercross':
+ bottomInfo.append(HT.BR(), HT.BR(), HT.Strong('Dominance Effect'),' is the difference between the mean trait value of cases heterozygous at a marker and the average mean for the two groups homozygous at this marker: e.g., BD - (BB+DD)/2]. A positive dominance effect indicates that the average phenotype of BD heterozygotes exceeds the mean of BB and DD homozygotes. No dominance deviation can be computed for a set of recombinant inbred strains or for a backcross.')
+ return rv,tblobj,bottomInfo
+
+ return rv,tblobj,bottomInfo
+
+ def plotIntMappingForPLINK(self, fd, canvas, offset= (80, 120, 20, 80), zoom = 1, startMb = None, endMb = None, showLocusForm = "",plinkResultDict={}):
+ #calculating margins
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ xLeftOffset = int(xLeftOffset*fontZoom)
+ xRightOffset = int(xRightOffset*fontZoom)
+ yBottomOffset = int(yBottomOffset*fontZoom)
+
+ cWidth = canvas.size[0]
+ cHeight = canvas.size[1]
+ plotWidth = cWidth - xLeftOffset - xRightOffset
+ plotHeight = cHeight - yTopOffset - yBottomOffset
+ startPixelX = xLeftOffset
+ endPixelX = (xLeftOffset + plotWidth)
+
+ #Drawing Area Height
+ drawAreaHeight = plotHeight
+ if self.plotScale == 'physic' and self.selectedChr > -1: # for single chr
+ drawAreaHeight -= self.ENSEMBL_BAND_HEIGHT + self.UCSC_BAND_HEIGHT+ self.WEBQTL_BAND_HEIGHT + 3*self.BAND_SPACING+ 10*zoom
+ if self.geneChecked:
+ drawAreaHeight -= self.NUM_GENE_ROWS*self.EACH_GENE_HEIGHT + 3*self.BAND_SPACING + 10*zoom
+ else:
+ if self.selectedChr > -1:
+ drawAreaHeight -= 20
+ else:# for all chrs
+ drawAreaHeight -= 30
+
+ #Image map
+ gifmap = HT.Map(name='WebQTLImageMap')
+
+ newoffset = (xLeftOffset, xRightOffset, yTopOffset, yBottomOffset)
+ # Draw the alternating-color background first and get plotXScale
+ plotXScale = self.drawGraphBackgroundForPLINK(canvas, gifmap, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb,plinkResultDict=plinkResultDict)
+
+ # Draw X axis
+ self.drawXAxisForPLINK(fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+ # Draw manhattam plot
+ self.drawManhattanPlotForPLINK(canvas, drawAreaHeight, gifmap, plotXScale, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb,plinkResultDict=plinkResultDict,thresholdPvalue=self.pValue)
+
+ return gifmap
+
+
+ def plotIntMapping(self, fd, canvas, offset= (80, 120, 20, 80), zoom = 1, startMb = None, endMb = None, showLocusForm = ""):
+ #calculating margins
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ xLeftOffset = int(xLeftOffset*fontZoom)
+ xRightOffset = int(xRightOffset*fontZoom)
+ yBottomOffset = int(yBottomOffset*fontZoom)
+
+ cWidth = canvas.size[0]
+ cHeight = canvas.size[1]
+ plotWidth = cWidth - xLeftOffset - xRightOffset
+ plotHeight = cHeight - yTopOffset - yBottomOffset
+ startPixelX = xLeftOffset
+ endPixelX = (xLeftOffset + plotWidth)
+
+ #Drawing Area Height
+ drawAreaHeight = plotHeight
+ if self.plotScale == 'physic' and self.selectedChr > -1: # for single chr
+ drawAreaHeight -= self.ENSEMBL_BAND_HEIGHT + self.UCSC_BAND_HEIGHT+ self.WEBQTL_BAND_HEIGHT + 3*self.BAND_SPACING+ 10*zoom
+ if self.geneChecked:
+ drawAreaHeight -= self.NUM_GENE_ROWS*self.EACH_GENE_HEIGHT + 3*self.BAND_SPACING + 10*zoom
+ else:# for all chrs
+ if self.selectedChr > -1:
+ drawAreaHeight -= 20
+ else:
+ drawAreaHeight -= 30
+
+ #Image map
+ gifmap = HT.Map(name='WebQTLImageMap')
+
+ newoffset = (xLeftOffset, xRightOffset, yTopOffset, yBottomOffset)
+ # Draw the alternating-color background first and get plotXScale
+ plotXScale = self.drawGraphBackground(canvas, gifmap, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+
+ # Draw X axis
+ self.drawXAxis(fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+ # Draw QTL curve
+ self.drawQTL(canvas, drawAreaHeight, gifmap, plotXScale, offset=newoffset, zoom= zoom, startMb=startMb, endMb = endMb)
+
+ #draw legend
+ if self.multipleInterval:
+ self.drawMultiTraitName(fd, canvas, gifmap, showLocusForm, offset=newoffset)
+ elif self.legendChecked:
+ self.drawLegendPanel(fd, canvas, offset=newoffset)
+ else:
+ pass
+
+ #draw position, no need to use a separate function
+ if fd.genotype.Mbmap:
+ self.drawProbeSetPosition(canvas, plotXScale, offset=newoffset)
+
+ return gifmap
+
+
+ # functions for manhattam plot of markers
+ def drawManhattanPlotForPLINK(self, canvas, drawAreaHeight, gifmap, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None,plinkResultDict={},thresholdPvalue=-1):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ # INTERCROSS = (self.genotype.type=="intercross")
+ INTERCROSS ='' #??????
+
+ ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+ GraphInterval=self.GraphInterval
+ pvalueHeightThresh = drawAreaHeight - 80 #ZS: Otherwise the plot gets very close to the chromosome labels
+
+ #draw the pvalue scale
+ #We first determine whether or not we are using a sliding scale.
+ #If so, we need to compute the maximum pvalue value to determine where the max y-value should be, and call this pvalueMax.
+ #pvalueTop is then defined to be above the pvalueMax by enough to add one additional pvalueScale increment.
+ #if we are using a set-scale, then we set pvalueTop to be the user's value, and pvalueMax doesn't matter.
+
+ # for human data we use p value instead of lrs
+ pValueList=[]
+ for key in plinkResultDict:
+ valueList = plinkResultDict[key]
+ for item in valueList:
+ pValue = item[-1]
+ pValueList.append(pValue)
+
+ formattedPValueList=[]
+ for pValue in pValueList:
+ try:
+ pValue=float(pValue)
+ except:
+ pValue =1
+ formattedpValue = -math.log10(pValue)
+ formattedPValueList.append(formattedpValue)
+
+ #sliding scale
+ pvalueMax = max(formattedPValueList)
+ #pvalueMax =pvalueMax +1
+ # no permutation result for plink func: GenReport()
+ pvalueMin = int(-math.log10(thresholdPvalue))
+
+ if pvalueMax> 100:
+ pvalueScale = 20.0
+ elif pvalueMax > 20:
+ pvalueScale = 5.0
+ elif pvalueMax > 7.5:
+ pvalueScale = 2.5
+ else:
+ pvalueScale = 1.0
+
+ # the base line for x-axis is -log(thresholdPvalue)
+ pvalueAxisList = Plot.frange(pvalueMin, pvalueMax, pvalueScale)
+ #make sure the user's value appears on the y-axis
+ #ZS: There is no way to do this without making the position of the points not directly proportional to a given distance on the y-axis
+ #tempPvalueMax=round(pvalueMax)
+ tempPvalueMax = pvalueAxisList[len(pvalueAxisList)-1] + pvalueScale
+ pvalueAxisList.append(tempPvalueMax)
+
+ #ZS: I don't understand this; the if statement will be true for any number that isn't exactly X.5.
+ #if abs(tempPvalueMax-pvalueMax) <0.5:
+ # tempPvalueMax=tempPvalueMax+1
+ # pvalueAxisList.append(tempPvalueMax)
+
+ #draw the "pvalue" string to the left of the axis
+ pvalueScaleFont=pid.Font(ttf="verdana", size=14*fontZoom, bold=0)
+ pvalueLODFont=pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ yZero = yTopOffset + plotHeight
+
+ #yAxis label display area
+ yAxis_label ='-log(P)'
+ canvas.drawString(yAxis_label, xLeftOffset - canvas.stringWidth("999.99", font=pvalueScaleFont) - 10*zoom, \
+ yZero - 150, font=pvalueLODFont, color=pid.black, angle=90)
+
+ for i,item in enumerate(pvalueAxisList):
+ ypvalue = yZero - (float(i)/float(len(pvalueAxisList) - 1)) * pvalueHeightThresh
+ canvas.drawLine(xLeftOffset, ypvalue, xLeftOffset - 4, ypvalue, color=self.LRS_COLOR, width=1*zoom)
+ scaleStr = "%2.1f" % item
+ #added by NL 6-24-2011:Y-axis scale display
+ canvas.drawString(scaleStr, xLeftOffset-4-canvas.stringWidth(scaleStr, font=pvalueScaleFont)-5, ypvalue+3, font=pvalueScaleFont, color=self.LRS_COLOR)
+
+ ChrList=self.ChrList
+ startPosX = xLeftOffset
+
+ for i, chr in enumerate(ChrList):
+
+ if plinkResultDict.has_key(chr):
+ plinkresultList = plinkResultDict[chr]
+
+ m = 0
+ #add by NL 06-24-2011: for mahanttam plot
+ symbolFont = pid.Font(ttf="fnt_bs", size=5,bold=0)
+ # color for point in each chr
+ chrCount=len(ChrList)
+ chrColorDict =self.getColorForMarker(chrCount=chrCount,flag=1)
+ for j, item in enumerate(plinkresultList):
+ try :
+ mb=float(item[1])/1000000.0
+ except:
+ mb=0
+
+ try :
+ pvalue =float(item[-1])
+ except:
+ pvalue =1
+
+ try:
+ snpName = item[0]
+ except:
+ snpName=''
+
+ formattedPvalue = -math.log10(pvalue)
+
+ Xc = startPosX + (mb-startMb)*plotXScale
+ Yc = yZero - (formattedPvalue-pvalueMin)*pvalueHeightThresh/(tempPvalueMax - pvalueMin)
+ canvas.drawString("5", Xc-canvas.stringWidth("5",font=symbolFont)/2+1,Yc+2,color=chrColorDict[i], font=symbolFont)
+ m += 1
+
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset, yTopOffset, color=self.LRS_COLOR, width=1*zoom) #the blue line running up the y axis
+
+ def drawQTL(self, canvas, drawAreaHeight, gifmap, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ INTERCROSS = (self.genotype.type=="intercross")
+
+ ChrLengthDistList = self.ChrLengthMbList
+ GraphInterval=self.GraphInterval
+ LRSHeightThresh = drawAreaHeight
+ AdditiveHeightThresh = drawAreaHeight/2
+ DominanceHeightThresh = drawAreaHeight/2
+
+ #draw the LRS scale
+ #We first determine whether or not we are using a sliding scale.
+ #If so, we need to compute the maximum LRS value to determine where the max y-value should be, and call this LRSMax.
+ #LRSTop is then defined to be above the LRSMax by enough to add one additional LRSScale increment.
+ #if we are using a set-scale, then we set LRSTop to be the user's value, and LRSMax doesn't matter.
+
+ if self.LRS_LOD == 'LOD':
+ lodm = self.LODFACTOR
+ else:
+ lodm = 1.0
+
+ if self.lrsMax <= 0: #sliding scale
+ LRSMax = max(map(max, self.qtlresults)).lrs
+ #genotype trait will give infinite LRS
+ LRSMax = min(LRSMax, webqtlConfig.MAXLRS)
+ LRSMax = max(self.significance, LRSMax)
+ else:
+ LRSMax = self.lrsMax*lodm
+
+ if LRSMax/lodm > 100:
+ LRSScale = 20.0
+ elif LRSMax/lodm > 20:
+ LRSScale = 5.0
+ elif LRSMax/lodm > 7.5:
+ LRSScale = 2.5
+ else:
+ LRSScale = 1.0
+
+ LRSAxisList = Plot.frange(LRSScale, LRSMax/lodm, LRSScale)
+ #make sure the user's value appears on the y-axis
+ #update by NL 6-21-2011: round the LOD value to 100 when LRSMax is equal to 460
+ LRSAxisList.append(round(LRSMax/lodm))
+
+ #draw the "LRS" or "LOD" string to the left of the axis
+ LRSScaleFont=pid.Font(ttf="verdana", size=14*fontZoom, bold=0)
+ LRSLODFont=pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ yZero = yTopOffset + plotHeight
+
+ #yAxis label display area
+ canvas.drawString(self.LRS_LOD, xLeftOffset - canvas.stringWidth("999.99", font=LRSScaleFont) - 10*zoom, \
+ yZero - 150, font=LRSLODFont, color=pid.black, angle=90)
+
+ for item in LRSAxisList:
+ yLRS = yZero - (item*lodm/LRSMax) * LRSHeightThresh
+ canvas.drawLine(xLeftOffset, yLRS, xLeftOffset - 4, yLRS, color=self.LRS_COLOR, width=1*zoom)
+ scaleStr = "%2.1f" % item
+ #added by NL 6-24-2011:Y-axis scale display
+ canvas.drawString(scaleStr, xLeftOffset-4-canvas.stringWidth(scaleStr, font=LRSScaleFont)-5, yLRS+3, font=LRSScaleFont, color=self.LRS_COLOR)
+
+
+ #"Significant" and "Suggestive" Drawing Routine
+ # ======= Draw the thick lines for "Significant" and "Suggestive" ===== (crowell: I tried to make the SNPs draw over these lines, but piddle wouldn't have it...)
+ if self.permChecked and not self.multipleInterval:
+ significantY = yZero - self.significance*LRSHeightThresh/LRSMax
+ suggestiveY = yZero - self.suggestive*LRSHeightThresh/LRSMax
+
+
+ startPosX = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ rightEdge = int(startPosX + self.ChrLengthDistList[i]*plotXScale - self.SUGGESTIVE_WIDTH/1.5)
+ #added by NL 6-24-2011:draw suggestive line (grey one)
+ canvas.drawLine(startPosX+self.SUGGESTIVE_WIDTH/1.5, suggestiveY, rightEdge, suggestiveY, color=self.SUGGESTIVE_COLOR,
+ width=self.SUGGESTIVE_WIDTH*zoom, clipX=(xLeftOffset, xLeftOffset + plotWidth-2))
+ #added by NL 6-24-2011:draw significant line (pink one)
+ canvas.drawLine(startPosX+self.SUGGESTIVE_WIDTH/1.5, significantY, rightEdge, significantY, color=self.SIGNIFICANT_COLOR,
+ width=self.SIGNIFICANT_WIDTH*zoom, clipX=(xLeftOffset, xLeftOffset + plotWidth-2))
+ sugg_coords = "%d, %d, %d, %d" % (startPosX, suggestiveY-2, rightEdge + 2*zoom, suggestiveY+2)
+ sig_coords = "%d, %d, %d, %d" % (startPosX, significantY-2, rightEdge + 2*zoom, significantY+2)
+ if self.LRS_LOD == 'LRS':
+ sugg_title = "Suggestive LRS = %0.2f" % self.suggestive
+ sig_title = "Significant LRS = %0.2f" % self.significance
+ else:
+ sugg_title = "Suggestive LOD = %0.2f" % (self.suggestive/4.61)
+ sig_title = "Significant LOD = %0.2f" % (self.significance/4.61)
+ Areas1 = HT.Area(shape='rect',coords=sugg_coords,title=sugg_title)
+ Areas2 = HT.Area(shape='rect',coords=sig_coords,title=sig_title)
+ gifmap.areas.append(Areas1)
+ gifmap.areas.append(Areas2)
+
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+
+ if self.multipleInterval:
+ lrsEdgeWidth = 1
+ else:
+ additiveMax = max(map(lambda X : abs(X.additive), self.qtlresults[0]))
+ if INTERCROSS:
+ dominanceMax = max(map(lambda X : abs(X.dominance), self.qtlresults[0]))
+ else:
+ dominanceMax = -1
+ lrsEdgeWidth = 2
+ for i, qtlresult in enumerate(self.qtlresults):
+ m = 0
+ startPosX = xLeftOffset
+ thisLRSColor = self.colorCollection[i]
+
+ #add by NL 06-24-2011: for mahanttam plot
+ symbolFont = pid.Font(ttf="fnt_bs", size=5,bold=0)
+
+ for j, _chr in enumerate(self.genotype):
+ chrCount=len(self.genotype)
+ chrColorDict =self.getColorForMarker(chrCount=chrCount,flag=1)
+ LRSCoordXY = []
+ AdditiveCoordXY = []
+ DominanceCoordXY = []
+ for k, _locus in enumerate(_chr):
+ if self.plotScale == 'physic':
+ Xc = startPosX + (_locus.Mb-startMb)*plotXScale
+ else:
+ Xc = startPosX + (_locus.cM-_chr[0].cM)*plotXScale
+ # updated by NL 06-18-2011:
+ # fix the over limit LRS graph issue since genotype trait may give infinite LRS;
+ # for any lrs is over than 460(LRS max in this system), it will be reset to 460
+ if qtlresult[m].lrs> 460 or qtlresult[m].lrs=='inf':
+ Yc = yZero - webqtlConfig.MAXLRS*LRSHeightThresh/LRSMax
+ else:
+ Yc = yZero - qtlresult[m].lrs*LRSHeightThresh/LRSMax
+
+ LRSCoordXY.append((Xc, Yc))
+ #add by NL 06-24-2011: for mahanttam plot
+ #self.significance/4.61 consider chr and LOD
+ # significantY = yZero - self.significance*LRSHeightThresh/LRSMax
+ # if Yc >significantY:
+ # canvas.drawString(":", Xc-canvas.stringWidth(":",font=symbolFont)/2+1,Yc+2,color=pid.black, font=symbolFont)
+ # else:
+ # canvas.drawString(":", Xc-canvas.stringWidth(":",font=symbolFont)/2+1,Yc+2,color=pid.black, font=symbolFont)
+
+ # add by NL 06-27-2011: eliminate imputed value when locus name is equal to '-'
+ if (qtlresult[m].locus.name) and (qtlresult[m].locus.name!=' - '):
+ canvas.drawString("5", Xc-canvas.stringWidth("5",font=symbolFont)/2+1,Yc+2,color=chrColorDict[j], font=symbolFont)
+
+ if not self.multipleInterval and self.additiveChecked:
+ Yc = yZero - qtlresult[m].additive*AdditiveHeightThresh/additiveMax
+ AdditiveCoordXY.append((Xc, Yc))
+ if not self.multipleInterval and INTERCROSS and self.additiveChecked:
+ Yc = yZero - qtlresult[m].dominance*DominanceHeightThresh/dominanceMax
+ DominanceCoordXY.append((Xc, Yc))
+ m += 1
+
+ startPosX += (ChrLengthDistList[j]+GraphInterval)*plotXScale
+
+
+ ###draw additive scale
+ if not self.multipleInterval and self.additiveChecked:
+ additiveScaleFont=pid.Font(ttf="verdana",size=12*fontZoom,bold=0)
+ additiveScale = Plot.detScaleOld(0,additiveMax)
+ additiveStep = (additiveScale[1]-additiveScale[0])/additiveScale[2]
+ additiveAxisList = Plot.frange(0, additiveScale[1], additiveStep)
+ maxAdd = additiveScale[1]
+ addPlotScale = AdditiveHeightThresh/additiveMax
+
+ additiveAxisList.append(additiveScale[1])
+ for item in additiveAxisList:
+ additiveY = yZero - item*addPlotScale
+ canvas.drawLine(xLeftOffset + plotWidth,additiveY,xLeftOffset+4+ plotWidth,additiveY,color=self.ADDITIVE_COLOR_POSITIVE, width=1*zoom)
+ scaleStr = "%2.3f" % item
+ canvas.drawString(scaleStr,xLeftOffset + plotWidth +6,additiveY+5,font=additiveScaleFont,color=self.ADDITIVE_COLOR_POSITIVE)
+
+ canvas.drawLine(xLeftOffset+plotWidth,additiveY,xLeftOffset+plotWidth,yZero,color=self.ADDITIVE_COLOR_POSITIVE, width=1*zoom)
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset, yTopOffset, color=self.LRS_COLOR, width=1*zoom) #the blue line running up the y axis
+
+ def drawGraphBackgroundForPLINK(self, canvas, gifmap, offset= (80, 120, 80, 50), zoom = 1, startMb = None, endMb = None,plinkResultDict={} ):
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #calculate plot scale
+ #XZ: all of these global variables should be passed from function signiture
+ ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+ GraphInterval=self.GraphInterval
+ ChrList =self.ChrList
+
+ #multiple chromosome view
+ plotXScale = plotWidth / ((len(ChrList)-1)*GraphInterval + drawRegionDistance)
+
+ startPosX = xLeftOffset
+ chrLabelFont=pid.Font(ttf="verdana",size=24*fontZoom,bold=0)
+
+ for i, _chr in enumerate(ChrList):
+
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+ # NL:resize chr width for drawing
+ if float(ChrLengthDistList[i])<90:
+ ChrLengthDistList[i]=90
+ #draw the shaded boxes and the sig/sug thick lines
+ canvas.drawRect(startPosX, yTopOffset, startPosX + ChrLengthDistList[i]*plotXScale, \
+ yTopOffset+plotHeight, edgeColor=pid.gainsboro,fillColor=theBackColor)
+
+ chrNameWidth = canvas.stringWidth(_chr, font=chrLabelFont)
+ chrStartPix = startPosX + (ChrLengthDistList[i]*plotXScale -chrNameWidth)/2
+ chrEndPix = startPosX + (ChrLengthDistList[i]*plotXScale +chrNameWidth)/2
+
+ canvas.drawString(_chr, chrStartPix, yTopOffset +20,font = chrLabelFont,color=pid.dimgray)
+ COORDS = "%d,%d,%d,%d" %(chrStartPix, yTopOffset, chrEndPix,yTopOffset +20)
+
+ #add by NL 09-03-2010
+ HREF = "javascript:changeView(%d,%s);" % (i,ChrLengthDistList)
+ Areas = HT.Area(shape='rect',coords=COORDS,href=HREF)
+ gifmap.areas.append(Areas)
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ return plotXScale
+
+
+ def drawGraphBackground(self, canvas, gifmap, offset= (80, 120, 80, 50), zoom = 1, startMb = None, endMb = None):
+ ##conditions
+ ##multiple Chromosome view
+ ##single Chromosome Physical
+ ##single Chromosome Genetic
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #calculate plot scale
+ if self.plotScale != 'physic':
+ self.ChrLengthDistList = self.ChrLengthCMList
+ drawRegionDistance = self.ChrLengthCMSum
+ else:
+ self.ChrLengthDistList = self.ChrLengthMbList
+ drawRegionDistance = self.ChrLengthMbSum
+
+ if self.selectedChr > -1: #single chromosome view
+ spacingAmt = plotWidth/13.5
+ i = 0
+ for startPix in Plot.frange(xLeftOffset, xLeftOffset+plotWidth, spacingAmt):
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+ i += 1
+ canvas.drawRect(startPix, yTopOffset, min(startPix+spacingAmt, xLeftOffset+plotWidth), \
+ yTopOffset+plotHeight, edgeColor=theBackColor, fillColor=theBackColor)
+
+ drawRegionDistance = self.ChrLengthDistList[self.selectedChr]
+ self.ChrLengthDistList = [drawRegionDistance]
+ if self.plotScale == 'physic':
+ plotXScale = plotWidth / (endMb-startMb)
+ else:
+ plotXScale = plotWidth / drawRegionDistance
+
+ else: #multiple chromosome view
+ plotXScale = plotWidth / ((len(self.genotype)-1)*self.GraphInterval + drawRegionDistance)
+
+ startPosX = xLeftOffset
+ chrLabelFont=pid.Font(ttf="verdana",size=24*fontZoom,bold=0)
+
+ for i, _chr in enumerate(self.genotype):
+
+ if (i % 2 == 0):
+ theBackColor = self.GRAPH_BACK_DARK_COLOR
+ else:
+ theBackColor = self.GRAPH_BACK_LIGHT_COLOR
+
+ #draw the shaded boxes and the sig/sug thick lines
+ canvas.drawRect(startPosX, yTopOffset, startPosX + self.ChrLengthDistList[i]*plotXScale, \
+ yTopOffset+plotHeight, edgeColor=pid.gainsboro,fillColor=theBackColor)
+
+ chrNameWidth = canvas.stringWidth(_chr.name, font=chrLabelFont)
+ chrStartPix = startPosX + (self.ChrLengthDistList[i]*plotXScale -chrNameWidth)/2
+ chrEndPix = startPosX + (self.ChrLengthDistList[i]*plotXScale +chrNameWidth)/2
+
+ canvas.drawString(_chr.name, chrStartPix, yTopOffset +20,font = chrLabelFont,color=pid.dimgray)
+ COORDS = "%d,%d,%d,%d" %(chrStartPix, yTopOffset, chrEndPix,yTopOffset +20)
+
+ #add by NL 09-03-2010
+ HREF = "javascript:changeView(%d,%s);" % (i,self.ChrLengthMbList)
+ Areas = HT.Area(shape='rect',coords=COORDS,href=HREF)
+ gifmap.areas.append(Areas)
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+ return plotXScale
+
+ # XZ: The only difference of function drawXAxisForPLINK and function drawXAxis are the function name and the self.plotScale condition.
+ def drawXAxisForPLINK(self, fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #Parameters
+ ChrLengthDistList = self.ChrLengthMbList
+ GraphInterval=self.GraphInterval
+
+ NUM_MINOR_TICKS = 5 # Number of minor ticks between major ticks
+ X_MAJOR_TICK_THICKNESS = 2
+ X_MINOR_TICK_THICKNESS = 1
+ X_AXIS_THICKNESS = 1*zoom
+
+ # ======= Alex: Draw the X-axis labels (megabase location)
+ MBLabelFont = pid.Font(ttf="verdana", size=12*fontZoom, bold=0)
+ xMajorTickHeight = 15 # How high the tick extends below the axis
+ xMinorTickHeight = 5*zoom
+ xAxisTickMarkColor = pid.black
+ xAxisLabelColor = pid.black
+ fontHeight = 12*fontZoom # How tall the font that we're using is
+ spacingFromLabelToAxis = 20
+ spacingFromLineToLabel = 3
+
+ if self.plotScale == 'physic':
+ strYLoc = yZero + spacingFromLabelToAxis + canvas.fontHeight(MBLabelFont)
+ ###Physical single chromosome view
+ if self.selectedChr > -1:
+ graphMbWidth = endMb - startMb
+ XScale = Plot.detScale(startMb, endMb)
+ XStart, XEnd, XStep = XScale
+ if XStep < 8:
+ XStep *= 2
+ spacingAmtX = spacingAmt = (XEnd-XStart)/XStep
+
+ j = 0
+ while abs(spacingAmtX -int(spacingAmtX)) >= spacingAmtX/100.0 and j < 6:
+ j += 1
+ spacingAmtX *= 10
+
+ formatStr = '%%2.%df' % j
+
+ for counter, _Mb in enumerate(Plot.frange(XStart, XEnd, spacingAmt / NUM_MINOR_TICKS)):
+ if _Mb < startMb or _Mb > endMb:
+ continue
+ Xc = xLeftOffset + plotXScale*(_Mb - startMb)
+ if counter % NUM_MINOR_TICKS == 0: # Draw a MAJOR mark, not just a minor tick mark
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMajorTickHeight, color=xAxisTickMarkColor, width=X_MAJOR_TICK_THICKNESS) # Draw the MAJOR tick mark
+ labelStr = str(formatStr % _Mb) # What Mbase location to put on the label
+ strWidth = canvas.stringWidth(labelStr, font=MBLabelFont)
+ drawStringXc = (Xc - (strWidth / 2.0))
+ canvas.drawString(labelStr, drawStringXc, strYLoc, font=MBLabelFont, color=xAxisLabelColor, angle=0)
+ else:
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMinorTickHeight, color=xAxisTickMarkColor, width=X_MINOR_TICK_THICKNESS) # Draw the MINOR tick mark
+ # end else
+
+ ###Physical genome wide view
+ else:
+ distScale = 0
+ startPosX = xLeftOffset
+ for i, distLen in enumerate(ChrLengthDistList):
+ if distScale == 0: #universal scale in whole genome mapping
+ if distLen > 75:
+ distScale = 25
+ elif distLen > 30:
+ distScale = 10
+ else:
+ distScale = 5
+ for tickdists in range(distScale, ceil(distLen), distScale):
+ canvas.drawLine(startPosX + tickdists*plotXScale, yZero, startPosX + tickdists*plotXScale, yZero + 7, color=pid.black, width=1*zoom)
+ canvas.drawString(str(tickdists), startPosX+tickdists*plotXScale, yZero + 10*zoom, color=pid.black, font=MBLabelFont, angle=270)
+ startPosX += (ChrLengthDistList[i]+GraphInterval)*plotXScale
+
+ megabaseLabelFont = pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ canvas.drawString("Megabases", xLeftOffset + (plotWidth -canvas.stringWidth("Megabases", font=megabaseLabelFont))/2,
+ strYLoc + canvas.fontHeight(MBLabelFont) + 5*zoom, font=megabaseLabelFont, color=pid.black)
+ pass
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset+plotWidth, yZero, color=pid.black, width=X_AXIS_THICKNESS) # Draw the X axis itself
+
+ def drawXAxis(self, fd, canvas, drawAreaHeight, gifmap, plotXScale, showLocusForm, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ #Parameters
+ NUM_MINOR_TICKS = 5 # Number of minor ticks between major ticks
+ X_MAJOR_TICK_THICKNESS = 2
+ X_MINOR_TICK_THICKNESS = 1
+ X_AXIS_THICKNESS = 1*zoom
+
+ # ======= Alex: Draw the X-axis labels (megabase location)
+ MBLabelFont = pid.Font(ttf="verdana", size=12*fontZoom, bold=0)
+ xMajorTickHeight = 15 # How high the tick extends below the axis
+ xMinorTickHeight = 5*zoom
+ xAxisTickMarkColor = pid.black
+ xAxisLabelColor = pid.black
+ fontHeight = 12*fontZoom # How tall the font that we're using is
+ spacingFromLabelToAxis = 20
+ spacingFromLineToLabel = 3
+
+ if self.plotScale == 'physic':
+ strYLoc = yZero + spacingFromLabelToAxis + canvas.fontHeight(MBLabelFont)
+ ###Physical single chromosome view
+ if self.selectedChr > -1:
+ graphMbWidth = endMb - startMb
+ XScale = Plot.detScale(startMb, endMb)
+ XStart, XEnd, XStep = XScale
+ if XStep < 8:
+ XStep *= 2
+ spacingAmtX = spacingAmt = (XEnd-XStart)/XStep
+
+ j = 0
+ while abs(spacingAmtX -int(spacingAmtX)) >= spacingAmtX/100.0 and j < 6:
+ j += 1
+ spacingAmtX *= 10
+
+ formatStr = '%%2.%df' % j
+
+ for counter, _Mb in enumerate(Plot.frange(XStart, XEnd, spacingAmt / NUM_MINOR_TICKS)):
+ if _Mb < startMb or _Mb > endMb:
+ continue
+ Xc = xLeftOffset + plotXScale*(_Mb - startMb)
+ if counter % NUM_MINOR_TICKS == 0: # Draw a MAJOR mark, not just a minor tick mark
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMajorTickHeight, color=xAxisTickMarkColor, width=X_MAJOR_TICK_THICKNESS) # Draw the MAJOR tick mark
+ labelStr = str(formatStr % _Mb) # What Mbase location to put on the label
+ strWidth = canvas.stringWidth(labelStr, font=MBLabelFont)
+ drawStringXc = (Xc - (strWidth / 2.0))
+ canvas.drawString(labelStr, drawStringXc, strYLoc, font=MBLabelFont, color=xAxisLabelColor, angle=0)
+ else:
+ canvas.drawLine(Xc, yZero, Xc, yZero+xMinorTickHeight, color=xAxisTickMarkColor, width=X_MINOR_TICK_THICKNESS) # Draw the MINOR tick mark
+ # end else
+
+ ###Physical genome wide view
+ else:
+ distScale = 0
+ startPosX = xLeftOffset
+ for i, distLen in enumerate(self.ChrLengthDistList):
+ if distScale == 0: #universal scale in whole genome mapping
+ if distLen > 75:
+ distScale = 25
+ elif distLen > 30:
+ distScale = 10
+ else:
+ distScale = 5
+ for tickdists in range(distScale, ceil(distLen), distScale):
+ canvas.drawLine(startPosX + tickdists*plotXScale, yZero, startPosX + tickdists*plotXScale, yZero + 7, color=pid.black, width=1*zoom)
+ canvas.drawString(str(tickdists), startPosX+tickdists*plotXScale, yZero + 10*zoom, color=pid.black, font=MBLabelFont, angle=270)
+ startPosX += (self.ChrLengthDistList[i]+self.GraphInterval)*plotXScale
+
+ megabaseLabelFont = pid.Font(ttf="verdana", size=14*zoom*1.5, bold=0)
+ canvas.drawString("Megabases", xLeftOffset + (plotWidth -canvas.stringWidth("Megabases", font=megabaseLabelFont))/2,
+ strYLoc + canvas.fontHeight(MBLabelFont) + 5*zoom, font=megabaseLabelFont, color=pid.black)
+ pass
+ else:
+ ChrAInfo = []
+ preLpos = -1
+ distinctCount = 0.0
+ if len(self.genotype) > 1:
+ for i, _chr in enumerate(self.genotype):
+ thisChr = []
+ Locus0CM = _chr[0].cM
+ nLoci = len(_chr)
+ if nLoci <= 8:
+ for _locus in _chr:
+ if _locus.name != ' - ':
+ if _locus.cM != preLpos:
+ distinctCount += 1
+ preLpos = _locus.cM
+ thisChr.append([_locus.name, _locus.cM-Locus0CM])
+ else:
+ for j in (0, nLoci/4, nLoci/2, nLoci*3/4, -1):
+ while _chr[j].name == ' - ':
+ j += 1
+ if _chr[j].cM != preLpos:
+ distinctCount += 1
+ preLpos = _chr[j].cM
+ thisChr.append([_chr[j].name, _chr[j].cM-Locus0CM])
+ ChrAInfo.append(thisChr)
+ else:
+ for i, _chr in enumerate(self.genotype):
+ thisChr = []
+ Locus0CM = _chr[0].cM
+ for _locus in _chr:
+ if _locus.name != ' - ':
+ if _locus.cM != preLpos:
+ distinctCount += 1
+ preLpos = _locus.cM
+ thisChr.append([_locus.name, _locus.cM-Locus0CM])
+ ChrAInfo.append(thisChr)
+
+ stepA = (plotWidth+0.0)/distinctCount
+
+ LRectWidth = 10
+ LRectHeight = 3
+ offsetA = -stepA
+ lineColor = pid.lightblue
+ startPosX = xLeftOffset
+ for j, ChrInfo in enumerate(ChrAInfo):
+ preLpos = -1
+ for i, item in enumerate(ChrInfo):
+ Lname,Lpos = item
+ if Lpos != preLpos:
+ offsetA += stepA
+ differ = 1
+ else:
+ differ = 0
+ preLpos = Lpos
+ Lpos *= plotXScale
+ if self.selectedChr > -1:
+ Zorder = i % 5
+ else:
+ Zorder = 0
+ if differ:
+ canvas.drawLine(startPosX+Lpos,yZero,xLeftOffset+offsetA,\
+ yZero+25, color=lineColor)
+ canvas.drawLine(xLeftOffset+offsetA,yZero+25,xLeftOffset+offsetA,\
+ yZero+40+Zorder*(LRectWidth+3),color=lineColor)
+ rectColor = pid.orange
+ else:
+ canvas.drawLine(xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3)-3,\
+ xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3),color=lineColor)
+ rectColor = pid.deeppink
+ canvas.drawRect(xLeftOffset+offsetA, yZero+40+Zorder*(LRectWidth+3),\
+ xLeftOffset+offsetA-LRectHeight,yZero+40+Zorder*(LRectWidth+3)+LRectWidth,\
+ edgeColor=rectColor,fillColor=rectColor,edgeWidth = 0)
+ COORDS="%d,%d,%d,%d"%(xLeftOffset+offsetA-LRectHeight, yZero+40+Zorder*(LRectWidth+3),\
+ xLeftOffset+offsetA,yZero+40+Zorder*(LRectWidth+3)+LRectWidth)
+ HREF="javascript:showDatabase3('%s','%s','%s','');" % (showLocusForm,fd.RISet+"Geno", Lname)
+ Areas=HT.Area(shape='rect',coords=COORDS,href=HREF, title="Locus : " + Lname)
+ gifmap.areas.append(Areas)
+ ##piddle bug
+ if j == 0:
+ canvas.drawLine(startPosX,yZero,startPosX,yZero+40, color=lineColor)
+ startPosX += (self.ChrLengthDistList[j]+self.GraphInterval)*plotXScale
+
+ canvas.drawLine(xLeftOffset, yZero, xLeftOffset+plotWidth, yZero, color=pid.black, width=X_AXIS_THICKNESS) # Draw the X axis itself
+
+ def getColorForMarker(self, chrCount,flag):# no change is needed
+ chrColorDict={}
+ for i in range(chrCount):
+ if flag==1: # display blue and lightblue intercross
+ chrColorDict[i]=pid.black
+ elif flag==0:
+ if (i%2==0):
+ chrColorDict[i]=pid.blue
+ else:
+ chrColorDict[i]=pid.lightblue
+ else:#display different color for different chr
+ if i in [0,8,16]:
+ chrColorDict[i]=pid.black
+ elif i in [1,9,17]:
+ chrColorDict[i]=pid.red
+ elif i in [2,10,18]:
+ chrColorDict[i]=pid.lightgreen
+ elif i in [3,11,19]:
+ chrColorDict[i]=pid.blue
+ elif i in [4,12]:
+ chrColorDict[i]=pid.lightblue
+ elif i in [5,13]:
+ chrColorDict[i]=pid.hotpink
+ elif i in [6,14]:
+ chrColorDict[i]=pid.gold
+ elif i in [7,15]:
+ chrColorDict[i]=pid.grey
+
+ return chrColorDict
+
+
+ def drawProbeSetPosition(self, canvas, plotXScale, offset= (40, 120, 80, 10), zoom = 1, startMb = None, endMb = None):
+ if len(self.traitList) != 1:
+ return
+
+ xLeftOffset, xRightOffset, yTopOffset, yBottomOffset = offset
+ plotWidth = canvas.size[0] - xLeftOffset - xRightOffset
+ plotHeight = canvas.size[1] - yTopOffset - yBottomOffset
+ yZero = canvas.size[1] - yBottomOffset
+ fontZoom = zoom
+ if zoom == 2:
+ fontZoom = 1.5
+
+ try:
+ Chr = self.traitList[0].chr # self.traitListChr =self.traitList[0].chr=_vals need to change to chrList and mbList
+ Mb = self.traitList[0].mb # self.traitListMb =self.traitList[0].mb=_vals
+ except:
+ return
+
+ if self.plotScale == 'physic':
+ if self.selectedChr > -1:
+ if self.genotype[0].name != Chr or Mb < self.startMb or Mb > self.endMb:
+ return
+ else:
+ locPixel = xLeftOffset + (Mb-self.startMb)*plotXScale
+ else:
+ locPixel = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ if _chr.name != Chr:
+ locPixel += (self.ChrLengthDistList[i] + self.GraphInterval)*plotXScale
+ else:
+ locPixel += Mb*plotXScale
+ break
+ else:
+ if self.selectedChr > -1:
+ if self.genotype[0].name != Chr:
+ return
+ else:
+ for i, _locus in enumerate(self.genotype[0]):
+ #the trait's position is on the left of the first genotype
+ if i==0 and _locus.Mb >= Mb:
+ locPixel=-1
+ break
+
+ #the trait's position is between two traits
+ if i > 0 and self.genotype[0][i-1].Mb < Mb and _locus.Mb >= Mb:
+ locPixel = xLeftOffset + plotXScale*(self.genotype[0][i-1].cM+(_locus.cM-self.genotype[0][i-1].cM)*(Mb -self.genotype[0][i-1].Mb)/(_locus.Mb-self.genotype[0][i-1].Mb))
+ break
+
+ #the trait's position is on the right of the last genotype
+ if i==len(self.genotype[0]) and Mb>=_locus.Mb:
+ locPixel = -1
+ else:
+ locPixel = xLeftOffset
+ for i, _chr in enumerate(self.genotype):
+ if _chr.name != Chr:
+ locPixel += (self.ChrLengthDistList[i] + self.GraphInterval)*plotXScale
+ else:
+ locPixel += (Mb*(_chr[-1].cM-_chr[0].cM)/self.ChrLengthCMList[i])*plotXScale
+ break
+ if locPixel >= 0:
+ traitPixel = ((locPixel, yZero), (locPixel-6, yZero+12), (locPixel+6, yZero+12))
+ canvas.drawPolygon(traitPixel, edgeColor=pid.black, fillColor=self.TRANSCRIPT_LOCATION_COLOR, closed=1)
+
+ if self.legendChecked:
+ startPosY = 15
+ nCol = 2
+ smallLabelFont = pid.Font(ttf="trebuc", size=12, bold=1)
+ leftOffset = xLeftOffset+(nCol-1)*200
+ canvas.drawPolygon(((leftOffset+6, startPosY-6), (leftOffset, startPosY+6), (leftOffset+12, startPosY+6)), edgeColor=pid.black, fillColor=self.TRANSCRIPT_LOCATION_COLOR, closed=1)
+ canvas.drawString("Sequence Site", (leftOffset+15), (startPosY+5), smallLabelFont, self.TOP_RIGHT_INFO_COLOR)
+
+ # build dict based on plink result, key is chr, value is list of [snp,BP,pValue]
+ def getPlinkResultDict(self,outputFileName='',thresholdPvalue=-1,ChrOrderIdNameDict={}):
+
+ ChrList =self.ChrList
+ plinkResultDict={}
+
+ plinkResultfp = open("%s%s.qassoc"% (webqtlConfig.TMPDIR, outputFileName), "rb")
+
+ headerLine=plinkResultfp.readline()# read header line
+ line = plinkResultfp.readline()
+
+ valueList=[] # initialize value list, this list will include snp, bp and pvalue info
+ pValueList=[]
+ count=0
+
+ while line:
+ #convert line from str to list
+ lineList=self.buildLineList(line=line)
+
+ # only keep the records whose chromosome name is in db
+ if ChrOrderIdNameDict.has_key(int(lineList[0])) and lineList[-1] and lineList[-1].strip()!='NA':
+
+ chrName=ChrOrderIdNameDict[int(lineList[0])]
+ snp = lineList[1]
+ BP = lineList[2]
+ pValue = float(lineList[-1])
+ pValueList.append(pValue)
+
+ if plinkResultDict.has_key(chrName):
+ valueList=plinkResultDict[chrName]
+
+ # pvalue range is [0,1]
+ if thresholdPvalue >=0 and thresholdPvalue<=1:
+ if pValue < thresholdPvalue:
+ valueList.append((snp,BP,pValue))
+ count+=1
+
+ plinkResultDict[chrName]=valueList
+ valueList=[]
+ else:
+ if thresholdPvalue>=0 and thresholdPvalue<=1:
+ if pValue < thresholdPvalue:
+ valueList.append((snp,BP,pValue))
+ count+=1
+
+ if valueList:
+ plinkResultDict[chrName]=valueList
+
+ valueList=[]
+
+
+ line =plinkResultfp.readline()
+ else:
+ line=plinkResultfp.readline()
+
+ if pValueList:
+ minPvalue= min(pValueList)
+ else:
+ minPvalue=0
+
+ return count,minPvalue,plinkResultDict
+
+
+ ######################################################
+ # input: line: str,one line read from file
+ # function: convert line from str to list;
+ # output: lineList list
+ #######################################################
+ def buildLineList(self,line=None):
+
+ lineList = string.split(string.strip(line),' ')# irregular number of whitespaces between columns
+ lineList =[ item for item in lineList if item <>'']
+ lineList = map(string.strip, lineList)
+
+ return lineList
+
+ #added by NL: automatically generate pheno txt file for PLINK based on strainList passed from dataEditing page
+ def genPhenoTxtFileForPlink(self,phenoFileName='', RISetName='', probesetName='', valueDict={}):
+ pedFileStrainList=self.getStrainNameFromPedFile(RISetName=RISetName)
+ outputFile = open("%s%s.txt"%(webqtlConfig.TMPDIR,phenoFileName),"wb")
+ headerLine = 'FID\tIID\t%s\n'%probesetName
+ outputFile.write(headerLine)
+
+ newValueList=[]
+
+ #if valueDict does not include some strain, value will be set to -9999 as missing value
+ for item in pedFileStrainList:
+ try:
+ value=valueDict[item]
+ value=str(value).replace('value=','')
+ value=value.strip()
+ except:
+ value=-9999
+
+ newValueList.append(value)
+
+
+ newLine=''
+ for i, strain in enumerate(pedFileStrainList):
+ j=i+1
+ value=newValueList[i]
+ newLine+='%s\t%s\t%s\n'%(strain, strain, value)
+
+ if j%1000==0:
+ outputFile.write(newLine)
+ newLine=''
+
+ if newLine:
+ outputFile.write(newLine)
+
+ outputFile.close()
+
+ # get strain name from ped file in order
+ def getStrainNameFromPedFile(self, RISetName=''):
+ pedFileopen= open("%splink/%s.ped"%(webqtlConfig.HTMLPATH, RISetName),"r")
+ line =pedFileopen.readline()
+ strainNameList=[]
+
+ while line:
+ lineList=string.split(string.strip(line),'\t')
+ lineList=map(string.strip,lineList)
+
+ strainName=lineList[0]
+ strainNameList.append(strainName)
+
+ line =pedFileopen.readline()
+
+ return strainNameList
+
+ #################################################################
+ ## Generate Chr list, Chr OrderId and Retrieve Length Information
+ #################################################################
+ #def getChrNameOrderIdLength(self,RISet=''):
+ #
+ # try:
+ # query = """
+ # Select
+ # Chr_Length.Name,Chr_Length.OrderId,Length from Chr_Length, InbredSet
+ # where
+ # Chr_Length.SpeciesId = InbredSet.SpeciesId AND
+ # InbredSet.Name = '%s'
+ # Order by OrderId
+ # """ % (RISet)
+ # self.cursor.execute(query)
+ #
+ # results =self.cursor.fetchall()
+ # ChrList=[]
+ # ChrLengthMbList=[]
+ # ChrNameOrderIdDict={}
+ # ChrOrderIdNameDict={}
+ #
+ # for item in results:
+ # ChrList.append(item[0])
+ # ChrNameOrderIdDict[item[0]]=item[1] # key is chr name, value is orderId
+ # ChrOrderIdNameDict[item[1]]=item[0] # key is orderId, value is chr name
+ # ChrLengthMbList.append(item[2])
+ #
+ # except:
+ # ChrList=[]
+ # ChrNameOrderIdDict={}
+ # ChrLengthMbList=[]
+ #
+ # return ChrList,ChrNameOrderIdDict,ChrOrderIdNameDict,ChrLengthMbList
--
cgit v1.2.3
|