diff options
Diffstat (limited to 'wqflask/base/data_set.py')
| -rw-r--r-- | wqflask/base/data_set.py | 683 |
1 files changed, 331 insertions, 352 deletions
diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 4959457a..afffe780 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -19,6 +19,23 @@ # This module is used by GeneNetwork project (www.genenetwork.org) from __future__ import absolute_import, print_function, division +from db.call import fetchall, fetchone, fetch1 +from utility.logger import getLogger +from utility.tools import USE_GN_SERVER, USE_REDIS, flat_files, flat_file_exists, GN2_BASE_URL +from db.gn_server import menu_main +from pprint import pformat as pf +from MySQLdb import escape_string as escape +from maintenance import get_group_samplelists +from utility.tools import locate, locate_ignore_error, flat_files +from utility import gen_geno_ob +from utility import chunks +from utility.benchmark import Bench +from utility import webqtlUtil +from db import webqtlDatabaseFunction +from base import species +from base import webqtlConfig +import reaper +from flask import Flask, g import os import math import string @@ -26,45 +43,28 @@ import collections import codecs import json +import requests import gzip import cPickle as pickle import itertools -from operator import itemgetter from redis import Redis -Redis = Redis() -from flask import Flask, g +r = Redis() -import reaper - -from base import webqtlConfig -from base import species -from db import webqtlDatabaseFunction -from utility import webqtlUtil -from utility.benchmark import Bench -from utility import chunks -from utility.tools import locate, locate_ignore_error, flat_files - -from maintenance import get_group_samplelists - -from MySQLdb import escape_string as escape -from pprint import pformat as pf -from db.gn_server import menu_main -from db.call import fetchall,fetchone,fetch1 - -from utility.tools import USE_GN_SERVER, USE_REDIS, flat_files, flat_file_exists -from utility.logger import getLogger -logger = getLogger(__name__ ) +logger = getLogger(__name__) # Used by create_database to instantiate objects # Each subclass will add to this DS_NAME_MAP = {} -def create_dataset(dataset_name, dataset_type = None, get_samplelist = True, group_name = None): + +def create_dataset(dataset_name, dataset_type=None, get_samplelist=True, group_name=None): + if dataset_name == "Temp": + dataset_type = "Temp" + if not dataset_type: dataset_type = Dataset_Getter(dataset_name) - logger.debug("dataset_type", dataset_type) dataset_ob = DS_NAME_MAP[dataset_type] dataset_class = globals()[dataset_ob] @@ -73,9 +73,10 @@ def create_dataset(dataset_name, dataset_type = None, get_samplelist = True, gro else: return dataset_class(dataset_name, get_samplelist) -class Dataset_Types(object): - def __init__(self): +class DatasetType: + + def __init__(self, redis_instance): """Create a dictionary of samples where the value is set to Geno, Publish or ProbeSet. E.g. @@ -90,39 +91,93 @@ Publish or ProbeSet. E.g. 'B139_K_1206_R': 'ProbeSet' ... """ + self.redis_instance = redis_instance self.datasets = {} - if USE_GN_SERVER: - data = menu_main() - else: - file_name = "wqflask/static/new/javascript/dataset_menu_structure.json" - with open(file_name, 'r') as fh: - data = json.load(fh) - - for species in data['datasets']: - for group in data['datasets'][species]: - for dataset_type in data['datasets'][species][group]: - for dataset in data['datasets'][species][group][dataset_type]: - short_dataset_name = dataset[1] - if dataset_type == "Phenotypes": - new_type = "Publish" - elif dataset_type == "Genotypes": - new_type = "Geno" - else: - new_type = "ProbeSet" - self.datasets[short_dataset_name] = new_type - # Set LOG_LEVEL_DEBUG=5 to see the following: - logger.debugf(5,"datasets",self.datasets) + data = self.redis_instance.get("dataset_structure") + if data: + self.datasets = json.loads(data) + else: # ZS: I don't think this should ever run unless Redis is emptied + try: + data = json.loads(requests.get( + GN2_BASE_URL + "/api/v_pre1/gen_dropdown", timeout=5).content) + for species in data['datasets']: + for group in data['datasets'][species]: + for dataset_type in data['datasets'][species][group]: + for dataset in data['datasets'][species][group][dataset_type]: + short_dataset_name = dataset[1] + if dataset_type == "Phenotypes": + new_type = "Publish" + elif dataset_type == "Genotypes": + new_type = "Geno" + else: + new_type = "ProbeSet" + self.datasets[short_dataset_name] = new_type + except: + pass + + self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) + + def set_dataset_key(self, t, name): + """If name is not in the object's dataset dictionary, set it, and update + dataset_structure in Redis + + args: + t: Type of dataset structure which can be: 'mrna_expr', 'pheno', + 'other_pheno', 'geno' + name: The name of the key to inserted in the datasets dictionary + + """ + sql_query_mapping = { + 'mrna_expr': ("""SELECT ProbeSetFreeze.Id FROM """ + + """ProbeSetFreeze WHERE ProbeSetFreeze.Name = "{}" """), + 'pheno': ("""SELECT InfoFiles.GN_AccesionId """ + + """FROM InfoFiles, PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id AND """ + + """InfoFiles.InfoPageName = PublishFreeze.Name"""), + 'other_pheno': ("""SELECT PublishFreeze.Name """ + + """FROM PublishFreeze, InbredSet """ + + """WHERE InbredSet.Name = '{}' AND """ + + """PublishFreeze.InbredSetId = InbredSet.Id"""), + 'geno': ("""SELECT GenoFreeze.Id FROM GenoFreeze WHERE """ + + """GenoFreeze.Name = "{}" """) + } + + dataset_name_mapping = { + "mrna_expr": "ProbeSet", + "pheno": "Publish", + "other_pheno": "Publish", + "geno": "Geno", + } + + if t in ['pheno', 'other_pheno']: + name = name.replace("Publish", "") + + if bool(len(g.db.execute(sql_query_mapping[t].format(name)).fetchone())): + self.datasets[name] = dataset_name_mapping[t] + self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) + return True + + return None def __call__(self, name): - return self.datasets[name] + + if name not in self.datasets: + for t in ["mrna_expr", "pheno", "other_pheno", "geno"]: + # This has side-effects, with the end result being a truth-y value + if(self.set_dataset_key(t, name)): + break + return self.datasets.get(name, None) # Return None if name has not been set + # Do the intensive work at startup one time only -Dataset_Getter = Dataset_Types() +Dataset_Getter = DatasetType(r) + def create_datasets_list(): if USE_REDIS: key = "all_datasets" - result = Redis.get(key) + result = r.get(key) if result: logger.debug("Redis cache hit") @@ -138,16 +193,16 @@ def create_datasets_list(): for dataset_type in type_dict: query = "SELECT Name FROM {}".format(type_dict[dataset_type]) for result in fetchall(query): - #The query at the beginning of this function isn't - #necessary here, but still would rather just reuse - #it logger.debug("type: {}\tname: - #{}".format(dataset_type, result.Name)) + # The query at the beginning of this function isn't + # necessary here, but still would rather just reuse + # it logger.debug("type: {}\tname: + # {}".format(dataset_type, result.Name)) dataset = create_dataset(result.Name, dataset_type) datasets.append(dataset) if USE_REDIS: - Redis.set(key, pickle.dumps(datasets, pickle.HIGHEST_PROTOCOL)) - Redis.expire(key, 60*60) + r.set(key, pickle.dumps(datasets, pickle.HIGHEST_PROTOCOL)) + r.expire(key, 60*60) return datasets @@ -168,21 +223,33 @@ def mescape(*items): class Markers(object): """Todo: Build in cacheing so it saves us reading the same file more than once""" + def __init__(self, name): - json_data_fh = open(locate(name + '.json','genotype/json')) - try: - markers = json.load(json_data_fh) - except: - markers = [] + json_data_fh = open(locate(name + ".json", 'genotype/json')) + + markers = [] + with open("%s/%s_snps.txt" % (flat_files('genotype/bimbam'), name), 'r') as bimbam_fh: + if len(bimbam_fh.readline().split(", ")) > 2: + delimiter = ", " + elif len(bimbam_fh.readline().split(",")) > 2: + delimiter = "," + elif len(bimbam_fh.readline().split("\t")) > 2: + delimiter = "\t" + else: + delimiter = " " + for line in bimbam_fh: + marker = {} + marker['name'] = line.split(delimiter)[0].rstrip() + marker['Mb'] = float(line.split(delimiter)[1].rstrip())/1000000 + marker['chr'] = line.split(delimiter)[2].rstrip() + markers.append(marker) for marker in markers: - if (marker['chr'] != "X") and (marker['chr'] != "Y"): + if (marker['chr'] != "X") and (marker['chr'] != "Y") and (marker['chr'] != "M"): marker['chr'] = int(marker['chr']) marker['Mb'] = float(marker['Mb']) self.markers = markers - #logger.debug("self.markers:", self.markers) - def add_pvalues(self, p_values): logger.debug("length of self.markers:", len(self.markers)) @@ -190,7 +257,7 @@ class Markers(object): if type(p_values) is list: # THIS IS only needed for the case when we are limiting the number of p-values calculated - #if len(self.markers) > len(p_values): + # if len(self.markers) > len(p_values): # self.markers = self.markers[:len(p_values)] for marker, p_value in itertools.izip(self.markers, p_values): @@ -202,7 +269,7 @@ class Markers(object): marker['lrs_value'] = 0 else: marker['lod_score'] = -math.log10(marker['p_value']) - #Using -log(p) for the LRS; need to ask Rob how he wants to get LRS from p-values + # Using -log(p) for the LRS; need to ask Rob how he wants to get LRS from p-values marker['lrs_value'] = -math.log10(marker['p_value']) * 4.61 elif type(p_values) is dict: filtered_markers = [] @@ -217,18 +284,20 @@ class Markers(object): marker['lrs_value'] = 0 else: marker['lod_score'] = -math.log10(marker['p_value']) - #Using -log(p) for the LRS; need to ask Rob how he wants to get LRS from p-values - marker['lrs_value'] = -math.log10(marker['p_value']) * 4.61 + # Using -log(p) for the LRS; need to ask Rob how he wants to get LRS from p-values + marker['lrs_value'] = - \ + math.log10(marker['p_value']) * 4.61 filtered_markers.append(marker) - #else: + # else: #logger.debug("marker {} NOT in p_values".format(i)) - #self.markers.remove(marker) + # self.markers.remove(marker) #del self.markers[i] self.markers = filtered_markers + class HumanMarkers(Markers): - def __init__(self, name, specified_markers = []): + def __init__(self, name, specified_markers=[]): marker_data_fh = open(flat_files('mapping') + '/' + name + '.bim') self.markers = [] for line in marker_data_fh: @@ -251,7 +320,6 @@ class HumanMarkers(Markers): #logger.debug("markers is: ", pf(self.markers)) - def add_pvalues(self, p_values): super(HumanMarkers, self).add_pvalues(p_values) @@ -264,13 +332,15 @@ class DatasetGroup(object): has multiple datasets associated with it. """ + def __init__(self, dataset, name=None): """This sets self.group and self.group_id""" - #logger.debug("DATASET NAME2:", dataset.name) if name == None: - self.name, self.id, self.genetic_type = fetchone(dataset.query_for_group) + self.name, self.id, self.genetic_type = fetchone( + dataset.query_for_group) else: - self.name, self.id, self.genetic_type = fetchone("SELECT InbredSet.Name, InbredSet.Id, InbredSet.GeneticType FROM InbredSet where Name='%s'" % name) + self.name, self.id, self.genetic_type = fetchone( + "SELECT InbredSet.Name, InbredSet.Id, InbredSet.GeneticType FROM InbredSet where Name='%s'" % name) if self.name == 'BXD300': self.name = "BXD" @@ -278,7 +348,6 @@ class DatasetGroup(object): self.parlist = None self.get_f1_parent_strains() - self.accession_id = self.get_accession_id() self.mapping_id, self.mapping_names = self.get_mapping_methods() self.species = webqtlDatabaseFunction.retrieve_species(self.name) @@ -288,27 +357,16 @@ class DatasetGroup(object): self._datasets = None self.genofile = None - def get_accession_id(self): - results = g.db.execute("""select InfoFiles.GN_AccesionId from InfoFiles, PublishFreeze, InbredSet where - InbredSet.Name = %s and - PublishFreeze.InbredSetId = InbredSet.Id and - InfoFiles.InfoPageName = PublishFreeze.Name and - PublishFreeze.public > 0 and - PublishFreeze.confidentiality < 1 order by - PublishFreeze.CreateTime desc""", (self.name)).fetchone() - - if results != None: - return str(results[0]) - else: - return "None" - def get_mapping_methods(self): - mapping_id = g.db.execute("select MappingMethodId from InbredSet where Name= '%s'" % self.name).fetchone()[0] + mapping_id = g.db.execute( + "select MappingMethodId from InbredSet where Name= '%s'" % self.name).fetchone()[0] if mapping_id == "1": - mapping_names = ["QTLReaper", "PYLMM", "R/qtl"] + mapping_names = ["GEMMA", "QTLReaper", "R/qtl"] elif mapping_id == "2": mapping_names = ["GEMMA"] + elif mapping_id == "3": + mapping_names = ["R/qtl"] elif mapping_id == "4": mapping_names = ["GEMMA", "PLINK"] else: @@ -316,12 +374,7 @@ class DatasetGroup(object): return mapping_id, mapping_names - def get_specified_markers(self, markers = []): - self.markers = HumanMarkers(self.name, markers) - def get_markers(self): - logger.debug("self.species is:", self.species) - def check_plink_gemma(): if flat_file_exists("mapping"): MAPPING_PATH = flat_files("mapping")+"/" @@ -332,9 +385,12 @@ class DatasetGroup(object): if check_plink_gemma(): marker_class = HumanMarkers else: - marker_class = Markers + marker_class = Markers - self.markers = marker_class(self.name) + if self.genofile: + self.markers = marker_class(self.genofile[:-5]) + else: + self.markers = marker_class(self.name) def get_f1_parent_strains(self): try: @@ -348,33 +404,36 @@ class DatasetGroup(object): if maternal and paternal: self.parlist = [maternal, paternal] + def get_genofiles(self): + jsonfile = "%s/%s.json" % (webqtlConfig.GENODIR, self.name) + try: + f = open(jsonfile) + except: + return None + jsondata = json.load(f) + return jsondata['genofile'] + def get_samplelist(self): result = None - key = "samplelist:v2:" + self.name + key = "samplelist:v3:" + self.name if USE_REDIS: - result = Redis.get(key) + result = r.get(key) if result is not None: - #logger.debug("Sample List Cache hit!!!") - #logger.debug("Before unjsonifying {}: {}".format(type(result), result)) self.samplelist = json.loads(result) - #logger.debug(" type: ", type(self.samplelist)) - #logger.debug(" self.samplelist: ", self.samplelist) else: logger.debug("Cache not hit") - genotype_fn = locate_ignore_error(self.name+".geno",'genotype') - mapping_fn = locate_ignore_error(self.name+".fam",'mapping') - if mapping_fn: - self.samplelist = get_group_samplelists.get_samplelist("plink", mapping_fn) - elif genotype_fn: - self.samplelist = get_group_samplelists.get_samplelist("geno", genotype_fn) + genotype_fn = locate_ignore_error(self.name+".geno", 'genotype') + if genotype_fn: + self.samplelist = get_group_samplelists.get_samplelist( + "geno", genotype_fn) else: self.samplelist = None - logger.debug("Sample list: ",self.samplelist) + if USE_REDIS: - Redis.set(key, json.dumps(self.samplelist)) - Redis.expire(key, 60*5) + r.set(key, json.dumps(self.samplelist)) + r.expire(key, 60*5) def all_samples_ordered(self): result = [] @@ -382,26 +441,36 @@ class DatasetGroup(object): [result.extend(l) for l in lists if l] return result - def read_genotype_file(self): + def read_genotype_file(self, use_reaper=False): '''Read genotype from .geno file instead of database''' - #genotype_1 is Dataset Object without parents and f1 - #genotype_2 is Dataset Object with parents and f1 (not for intercross) + # genotype_1 is Dataset Object without parents and f1 + # genotype_2 is Dataset Object with parents and f1 (not for intercross) - genotype_1 = reaper.Dataset() + #genotype_1 = reaper.Dataset() # reaper barfs on unicode filenames, so here we ensure it's a string if self.genofile: - full_filename = str(locate(self.genofile, 'genotype')) + if "RData" in self.genofile: # ZS: This is a temporary fix; I need to change the way the JSON files that point to multiple genotype files are structured to point to other file types like RData + full_filename = str( + locate(self.genofile.split(".")[0] + ".geno", 'genotype')) + else: + full_filename = str(locate(self.genofile, 'genotype')) else: full_filename = str(locate(self.name + '.geno', 'genotype')) - genotype_1.read(full_filename) + + if use_reaper: + genotype_1 = reaper.Dataset() + genotype_1.read(full_filename) + else: + genotype_1 = gen_geno_ob.genotype(full_filename) if genotype_1.type == "group" and self.parlist: - genotype_2 = genotype_1.add(Mat=self.parlist[0], Pat=self.parlist[1]) #, F1=_f1) + genotype_2 = genotype_1.add( + Mat=self.parlist[0], Pat=self.parlist[1]) # , F1=_f1) else: genotype_2 = genotype_1 - #determine default genotype object + # determine default genotype object if self.incparentsf1 and genotype_1.type != "intercross": genotype = genotype_2 else: @@ -412,24 +481,21 @@ class DatasetGroup(object): return genotype -def datasets(group_name, this_group = None): + +def datasets(group_name, this_group=None): key = "group_dataset_menu:v2:" + group_name - logger.debug("key is2:", key) dataset_menu = [] - logger.debug("[tape4] webqtlConfig.PUBLICTHRESH:", webqtlConfig.PUBLICTHRESH) - logger.debug("[tape4] type webqtlConfig.PUBLICTHRESH:", type(webqtlConfig.PUBLICTHRESH)) the_results = fetchall(''' (SELECT '#PublishFreeze',PublishFreeze.FullName,PublishFreeze.Name FROM PublishFreeze,InbredSet WHERE PublishFreeze.InbredSetId = InbredSet.Id and InbredSet.Name = '%s' - and PublishFreeze.public > %s) + ORDER BY PublishFreeze.Id ASC) UNION (SELECT '#GenoFreeze',GenoFreeze.FullName,GenoFreeze.Name FROM GenoFreeze, InbredSet WHERE GenoFreeze.InbredSetId = InbredSet.Id - and InbredSet.Name = '%s' - and GenoFreeze.public > %s) + and InbredSet.Name = '%s') UNION (SELECT Tissue.Name, ProbeSetFreeze.FullName,ProbeSetFreeze.Name FROM ProbeSetFreeze, ProbeFreeze, InbredSet, Tissue @@ -437,28 +503,37 @@ def datasets(group_name, this_group = None): and ProbeFreeze.TissueId = Tissue.Id and ProbeFreeze.InbredSetId = InbredSet.Id and InbredSet.Name like %s - and ProbeSetFreeze.public > %s - ORDER BY Tissue.Name, ProbeSetFreeze.CreateTime desc, ProbeSetFreeze.AvgId) - ''' % (group_name, webqtlConfig.PUBLICTHRESH, - group_name, webqtlConfig.PUBLICTHRESH, - "'" + group_name + "'", webqtlConfig.PUBLICTHRESH)) - - #for tissue_name, dataset in itertools.groupby(the_results, itemgetter(0)): - for dataset_item in the_results: + ORDER BY Tissue.Name, ProbeSetFreeze.OrderList DESC) + ''' % (group_name, + group_name, + "'" + group_name + "'")) + + sorted_results = sorted(the_results, key=lambda kv: kv[0]) + + # ZS: This is kind of awkward, but need to ensure Phenotypes show up before Genotypes in dropdown + pheno_inserted = False + geno_inserted = False + for dataset_item in sorted_results: tissue_name = dataset_item[0] dataset = dataset_item[1] dataset_short = dataset_item[2] if tissue_name in ['#PublishFreeze', '#GenoFreeze']: - dataset_menu.append(dict(tissue=None, datasets=[(dataset, dataset_short)])) + if tissue_name == '#PublishFreeze' and (dataset_short == group_name + 'Publish'): + dataset_menu.insert( + 0, dict(tissue=None, datasets=[(dataset, dataset_short)])) + pheno_inserted = True + elif pheno_inserted and tissue_name == '#GenoFreeze': + dataset_menu.insert( + 1, dict(tissue=None, datasets=[(dataset, dataset_short)])) + geno_inserted = True + else: + dataset_menu.append( + dict(tissue=None, datasets=[(dataset, dataset_short)])) else: - dataset_sub_menu = [item[1:] for item in dataset] - tissue_already_exists = False - tissue_position = None for i, tissue_dict in enumerate(dataset_menu): if tissue_dict['tissue'] == tissue_name: tissue_already_exists = True - tissue_position = i break if tissue_already_exists: @@ -466,11 +541,11 @@ def datasets(group_name, this_group = None): dataset_menu[i]['datasets'].append((dataset, dataset_short)) else: dataset_menu.append(dict(tissue=tissue_name, - datasets=[(dataset, dataset_short)])) + datasets=[(dataset, dataset_short)])) if USE_REDIS: - Redis.set(key, pickle.dumps(dataset_menu, pickle.HIGHEST_PROTOCOL)) - Redis.expire(key, 60*5) + r.set(key, pickle.dumps(dataset_menu, pickle.HIGHEST_PROTOCOL)) + r.expire(key, 60*5) if this_group != None: this_group._datasets = dataset_menu @@ -478,6 +553,7 @@ def datasets(group_name, this_group = None): else: return dataset_menu + class DataSet(object): """ DataSet class defines a dataset in webqtl, can be either Microarray, @@ -485,7 +561,7 @@ class DataSet(object): """ - def __init__(self, name, get_samplelist = True, group_name = None): + def __init__(self, name, get_samplelist=True, group_name=None): assert name, "Need a name" self.name = name @@ -493,21 +569,23 @@ class DataSet(object): self.shortname = None self.fullname = None self.type = None - self.data_scale = None #ZS: For example log2 + self.data_scale = None # ZS: For example log2 self.setup() - if self.type == "Temp": #Need to supply group name as input if temp trait - self.group = DatasetGroup(self, group_name) # sets self.group and self.group_id and gets genotype + if self.type == "Temp": # Need to supply group name as input if temp trait + # sets self.group and self.group_id and gets genotype + self.group = DatasetGroup(self, name=group_name) else: self.check_confidentiality() self.retrieve_other_names() - self.group = DatasetGroup(self) # sets self.group and self.group_id and gets genotype + # sets self.group and self.group_id and gets genotype + self.group = DatasetGroup(self) + self.accession_id = self.get_accession_id() if get_samplelist == True: - self.group.get_samplelist() + self.group.get_samplelist() self.species = species.TheSpecies(self) - def get_desc(self): """Gets overridden later, at least for Temp...used by trait's get_given_name""" return None @@ -517,6 +595,31 @@ class DataSet(object): def riset(): Weve_Renamed_This_As_Group + def get_accession_id(self): + if self.type == "Publish": + results = g.db.execute("""select InfoFiles.GN_AccesionId from InfoFiles, PublishFreeze, InbredSet where + InbredSet.Name = %s and + PublishFreeze.InbredSetId = InbredSet.Id and + InfoFiles.InfoPageName = PublishFreeze.Name and + PublishFreeze.public > 0 and + PublishFreeze.confidentiality < 1 order by + PublishFreeze.CreateTime desc""", (self.group.name)).fetchone() + elif self.type == "Geno": + results = g.db.execute("""select InfoFiles.GN_AccesionId from InfoFiles, GenoFreeze, InbredSet where + InbredSet.Name = %s and + GenoFreeze.InbredSetId = InbredSet.Id and + InfoFiles.InfoPageName = GenoFreeze.ShortName and + GenoFreeze.public > 0 and + GenoFreeze.confidentiality < 1 order by + GenoFreeze.CreateTime desc""", (self.group.name)).fetchone() + else: + results = None + + if results != None: + return str(results[0]) + else: + return "None" + def retrieve_other_names(self): """This method fetches the the dataset names in search_result. @@ -529,29 +632,26 @@ class DataSet(object): """ - try: if self.type == "ProbeSet": query_args = tuple(escape(x) for x in ( - str(webqtlConfig.PUBLICTHRESH), self.name, self.name, self.name)) self.id, self.name, self.fullname, self.shortname, self.data_scale, self.tissue = fetch1(""" -SELECT ProbeSetFreeze.Id, ProbeSetFreeze.Name, ProbeSetFreeze.FullName, ProbeSetFreeze.ShortName, ProbeSetFreeze.DataScale, Tissue.Name -FROM ProbeSetFreeze, ProbeFreeze, Tissue -WHERE ProbeSetFreeze.public > %s -AND ProbeSetFreeze.ProbeFreezeId = ProbeFreeze.Id -AND ProbeFreeze.TissueId = Tissue.Id -AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFreeze.ShortName = '%s') - """ % (query_args),"/dataset/"+self.name+".json", - lambda r: (r["id"],r["name"],r["full_name"],r["short_name"],r["data_scale"],r["tissue"]) + SELECT ProbeSetFreeze.Id, ProbeSetFreeze.Name, ProbeSetFreeze.FullName, ProbeSetFreeze.ShortName, ProbeSetFreeze.DataScale, Tissue.Name + FROM ProbeSetFreeze, ProbeFreeze, Tissue + WHERE ProbeSetFreeze.ProbeFreezeId = ProbeFreeze.Id + AND ProbeFreeze.TissueId = Tissue.Id + AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFreeze.ShortName = '%s') + """ % (query_args), "/dataset/"+self.name+".json", + lambda r: (r["id"], r["name"], r["full_name"], + r["short_name"], r["data_scale"], r["tissue"]) ) else: query_args = tuple(escape(x) for x in ( (self.type + "Freeze"), - str(webqtlConfig.PUBLICTHRESH), self.name, self.name, self.name)) @@ -560,12 +660,12 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre self.id, self.name, self.fullname, self.shortname = fetchone(""" SELECT Id, Name, FullName, ShortName FROM %s - WHERE public > %s AND - (Name = '%s' OR FullName = '%s' OR ShortName = '%s') - """ % (query_args)) + WHERE (Name = '%s' OR FullName = '%s' OR ShortName = '%s') + """ % (query_args)) except TypeError: - logger.debug("Dataset {} is not yet available in GeneNetwork.".format(self.name)) + logger.debug( + "Dataset {} is not yet available in GeneNetwork.".format(self.name)) pass def get_trait_data(self, sample_list=None): @@ -623,7 +723,7 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre and {}.Id = {}XRef.{}Id order by {}.Id """.format(*mescape(self.type, self.type, self.type, self.name, - dataset_type, self.type, dataset_type, dataset_type)) + dataset_type, self.type, dataset_type, dataset_type)) else: query += """ WHERE {}XRef.{}FreezeId = {}Freeze.Id @@ -631,7 +731,7 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre and {}.Id = {}XRef.{}Id order by {}.Id """.format(*mescape(self.type, self.type, self.type, self.type, - self.name, dataset_type, self.type, self.type, dataset_type)) + self.name, dataset_type, self.type, self.type, dataset_type)) #logger.debug("trait data query: ", query) @@ -651,6 +751,7 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre self.trait_data[trait_name] += ( trait_sample_data[chunk_counter][trait_counter][data_start_pos:]) + class PhenotypeDataSet(DataSet): DS_NAME_MAP['Publish'] = 'PhenotypeDataSet' @@ -660,24 +761,26 @@ class PhenotypeDataSet(DataSet): # 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'] + 'Phenotype.Pre_publication_description', + 'Phenotype.Pre_publication_abbreviation', + 'Phenotype.Post_publication_abbreviation', + 'PublishXRef.mean', + 'Phenotype.Lab_code', + 'Publication.PubMed_ID', + 'Publication.Abstract', + 'Publication.Title', + 'Publication.Authors', + 'PublishXRef.Id'] # Figure out what display_fields is - self.display_fields = ['name', + self.display_fields = ['name', 'group_code', 'pubmed_id', 'pre_publication_description', 'post_publication_description', 'original_description', 'pre_publication_abbreviation', 'post_publication_abbreviation', + 'mean', 'lab_code', 'submitter', 'owner', 'authorized_users', @@ -689,13 +792,13 @@ class PhenotypeDataSet(DataSet): # Fields displayed in the search results table header self.header_fields = ['Index', - 'Record', - 'Description', - 'Authors', - 'Year', - 'Max LRS', - 'Max LRS Location', - 'Additive Effect'] + 'Record', + 'Description', + 'Authors', + 'Year', + 'Max LRS', + 'Max LRS Location', + 'Additive Effect'] self.type = 'Publish' @@ -713,21 +816,7 @@ class PhenotypeDataSet(DataSet): # (Urgently?) Need to write this pass - def get_trait_list(self): - query = """ - select PublishXRef.Id - from PublishXRef, PublishFreeze - where PublishFreeze.InbredSetId=PublishXRef.InbredSetId - and PublishFreeze.Id = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - - def get_trait_info(self, trait_list, species = ''): + def get_trait_info(self, trait_list, species=''): for this_trait in trait_list: if not this_trait.haveinfo: @@ -735,12 +824,12 @@ class PhenotypeDataSet(DataSet): description = this_trait.post_publication_description - #If the dataset is confidential and the user has access to confidential - #phenotype traits, then display the pre-publication description instead - #of the post-publication description + # If the dataset is confidential and the user has access to confidential + # phenotype traits, then display the pre-publication description instead + # of the post-publication description if this_trait.confidential: this_trait.description_display = "" - continue # for now + continue # for now, because no authorization features if not webqtlUtil.hasAccessToConfidentialPhenotypeTrait( privilege=self.privilege, @@ -762,11 +851,9 @@ class PhenotypeDataSet(DataSet): if this_trait.pubmed_id: this_trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % this_trait.pubmed_id - #LRS and its location + # 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: query = """ @@ -783,18 +870,9 @@ class PhenotypeDataSet(DataSet): 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: %.6f' % (LRS_Chr, float(LRS_Mb)) + this_trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % ( + LRS_Chr, float(LRS_Mb)) def retrieve_sample_data(self, trait): query = """ @@ -855,45 +933,19 @@ class GenotypeDataSet(DataSet): def check_confidentiality(self): return geno_mrna_confidentiality(self) - def get_trait_list(self): - query = """ - select Geno.Name - from Geno, GenoXRef - where GenoXRef.GenoId = Geno.Id - and GenoFreezeId = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - def get_trait_info(self, trait_list, species=None): 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: %.6f' % (this_trait.chr, float(this_trait.mb) ) - this_trait.location_value = trait_location_value + this_trait.location_repr = 'Chr%s: %.6f' % ( + this_trait.chr, float(this_trait.mb)) def retrieve_sample_data(self, trait): query = """ SELECT - Strain.Name, GenoData.value, GenoSE.error, GenoData.Id, Strain.Name2 + Strain.Name, GenoData.value, GenoSE.error, "N/A", Strain.Name2 FROM (GenoData, GenoFreeze, Strain, Geno, GenoXRef) left join GenoSE on @@ -945,6 +997,7 @@ class MrnaAssayDataSet(DataSet): 'blatseq', 'targetseq', 'chipid', 'comments', 'strand_probe', 'strand_gene', + 'proteinid', 'uniprotid', 'probe_set_target_region', 'probe_set_specificity', 'probe_set_blat_score', @@ -956,14 +1009,14 @@ class MrnaAssayDataSet(DataSet): # Fields displayed in the search results table header self.header_fields = ['Index', - 'Record', - 'Symbol', - 'Description', - 'Location', - 'Mean', - 'Max LRS', - 'Max LRS Location', - 'Additive Effect'] + 'Record', + 'Symbol', + 'Description', + 'Location', + 'Mean', + 'Max LRS', + 'Max LRS Location', + 'Additive Effect'] # Todo: Obsolete or rename this field self.type = 'ProbeSet' @@ -979,24 +1032,9 @@ class MrnaAssayDataSet(DataSet): ProbeSetFreeze.Name = "%s" ''' % escape(self.name) - def check_confidentiality(self): return geno_mrna_confidentiality(self) - def get_trait_list_1(self): - query = """ - select ProbeSet.Name - from ProbeSet, ProbeSetXRef - where ProbeSetXRef.ProbeSetId = ProbeSet.Id - and ProbeSetFreezeId = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - def get_trait_info(self, trait_list=None, species=''): # Note: setting trait_list to [] is probably not a great idea. @@ -1011,10 +1049,12 @@ class MrnaAssayDataSet(DataSet): if not this_trait.symbol: this_trait.symbol = "N/A" - #XZ, 12/08/2008: description - #XZ, 06/05/2009: Rob asked to add probe target description - description_string = unicode(str(this_trait.description).strip(codecs.BOM_UTF8), 'utf-8') - target_string = unicode(str(this_trait.probe_target_description).strip(codecs.BOM_UTF8), 'utf-8') + # XZ, 12/08/2008: description + # XZ, 06/05/2009: Rob asked to add probe target description + description_string = unicode( + str(this_trait.description).strip(codecs.BOM_UTF8), 'utf-8') + target_string = unicode( + str(this_trait.probe_target_description).strip(codecs.BOM_UTF8), 'utf-8') if len(description_string) > 1 and description_string != 'None': description_display = description_string @@ -1028,31 +1068,13 @@ class MrnaAssayDataSet(DataSet): # Save it for the jinja2 template 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: - #Checks if the chromosome number can be cast to an int (i.e. isn't "X" or "Y") - #This is so we can convert the location to a number used for sorting - trait_location_value = self.convert_location_to_value(this_trait.chr, this_trait.mb) - #try: - # trait_location_value = int(this_trait.chr)*1000 + this_trait.mb - #except ValueError: - # 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) - - #ZS: Put this in function currently called "convert_location_to_value" - this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, - float(this_trait.mb)) - this_trait.location_value = trait_location_value - - #Get mean expression value + this_trait.location_repr = 'Chr%s: %.6f' % ( + this_trait.chr, float(this_trait.mb)) + + # Get mean expression value query = ( - """select ProbeSetXRef.mean from ProbeSetXRef, ProbeSet + """select ProbeSetXRef.mean from ProbeSetXRef, ProbeSet where ProbeSetXRef.ProbeSetFreezeId = %s and ProbeSet.Id = ProbeSetXRef.ProbeSetId and ProbeSet.Name = '%s' @@ -1068,13 +1090,11 @@ class MrnaAssayDataSet(DataSet): if mean: this_trait.mean = "%2.3f" % mean - #LRS and its location + # 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 + # Max LRS and its Locus location if this_trait.lrs and this_trait.locus: query = """ select Geno.Chr, Geno.Mb from Geno, Species @@ -1087,49 +1107,23 @@ class MrnaAssayDataSet(DataSet): if result: lrs_chr, lrs_mb = result - #XZ: LRS_location_value is used for sorting - lrs_location_value = self.convert_location_to_value(lrs_chr, lrs_mb) this_trait.LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_score_value = this_trait.lrs - this_trait.LRS_location_repr = 'Chr%s: %.6f' % (lrs_chr, float(lrs_mb)) - - - def convert_location_to_value(self, chromosome, mb): - try: - location_value = int(chromosome)*1000 + float(mb) - except ValueError: - if chromosome.upper() == 'X': - location_value = 20*1000 + float(mb) - else: - location_value = (ord(str(chromosome).upper()[0])*1000 + - float(mb)) - - return location_value + this_trait.LRS_location_repr = 'Chr%s: %.6f' % ( + lrs_chr, float(lrs_mb)) - def get_sequence(self): - query = """ - SELECT - ProbeSet.BlatSeq - FROM - ProbeSet, ProbeSetFreeze, ProbeSetXRef - WHERE - ProbeSet.Id=ProbeSetXRef.ProbeSetId and - ProbeSetFreeze.Id = ProbeSetXRef.ProbSetFreezeId and - ProbeSet.Name = %s - ProbeSetFreeze.Name = %s - """ % (escape(self.name), escape(self.dataset.name)) - logger.sql(query) - results = g.db.execute(query).fetchone() - return results[0] + return trait_list def retrieve_sample_data(self, trait): query = """ SELECT - Strain.Name, ProbeSetData.value, ProbeSetSE.error, ProbeSetData.Id, Strain.Name2 + Strain.Name, ProbeSetData.value, ProbeSetSE.error, NStrain.count, Strain.Name2 FROM (ProbeSetData, ProbeSetFreeze, Strain, ProbeSet, ProbeSetXRef) left join ProbeSetSE on (ProbeSetSE.DataId = ProbeSetData.Id AND ProbeSetSE.StrainId = ProbeSetData.StrainId) + left join NStrain on + (NStrain.DataId = ProbeSetData.Id AND + NStrain.StrainId = ProbeSetData.StrainId) WHERE ProbeSet.Name = '%s' AND ProbeSetXRef.ProbeSetId = ProbeSet.Id AND ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND @@ -1144,7 +1138,6 @@ class MrnaAssayDataSet(DataSet): #logger.debug("RETRIEVED RESULTS HERE:", results) return results - def retrieve_genes(self, column_name): query = """ select ProbeSet.Name, ProbeSet.%s @@ -1180,7 +1173,6 @@ class TempDataSet(DataSet): self.fullname = 'Temporary Storage' self.shortname = 'Temp' - @staticmethod def handle_pca(desc): if 'PCA' in desc: @@ -1198,19 +1190,6 @@ class TempDataSet(DataSet): desc = self.handle_pca(desc) return desc - def get_group(self): - query = """ - SELECT - InbredSet.Name, InbredSet.Id - FROM - InbredSet, Temp - WHERE - Temp.InbredSetId = InbredSet.Id AND - Temp.Name = "%s" - """ % self.name - logger.sql(query) - self.group, self.group_id = g.db.execute(query).fetchone() - def retrieve_sample_data(self, trait): query = """ SELECT @@ -1234,7 +1213,7 @@ def geno_mrna_confidentiality(ob): #logger.debug("dataset_table [%s]: %s" % (type(dataset_table), dataset_table)) query = '''SELECT Id, Name, FullName, confidentiality, - AuthorisedUsers FROM %s WHERE Name = "%s"''' % (dataset_table,ob.name) + AuthorisedUsers FROM %s WHERE Name = "%s"''' % (dataset_table, ob.name) logger.sql(query) result = g.db.execute(query) |