diff options
Diffstat (limited to 'wqflask/base')
| -rwxr-xr-x | wqflask/base/data_set.py | 556 | ||||
| -rwxr-xr-x | wqflask/base/webqtlTrait.py | 25 |
2 files changed, 467 insertions, 114 deletions
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': |