diff options
| author | Arthur Centeno | 2021-10-25 21:04:23 +0000 |
|---|---|---|
| committer | Arthur Centeno | 2021-10-25 21:04:23 +0000 |
| commit | 499a80f138030c4de1629c043c8f9401a99894ea (patch) | |
| tree | 449dcae965d13f966fb6d52625fbc86661c8c6a0 /wqflask/base | |
| parent | 6151faa9ea67af4bf4ea95fb681a9dc4319474b6 (diff) | |
| parent | 700802303e5e8221a9d591ba985d6607aa61e1ce (diff) | |
| download | genenetwork2-499a80f138030c4de1629c043c8f9401a99894ea.tar.gz | |
Merge github.com:genenetwork/genenetwork2 into acenteno
Diffstat (limited to 'wqflask/base')
| -rw-r--r-- | wqflask/base/GeneralObject.py | 78 | ||||
| -rw-r--r-- | wqflask/base/data_set.py | 683 | ||||
| -rw-r--r-- | wqflask/base/mrna_assay_tissue_data.py | 45 | ||||
| -rw-r--r-- | wqflask/base/species.py | 96 | ||||
| -rw-r--r-- | wqflask/base/trait.py | 501 | ||||
| -rw-r--r-- | wqflask/base/webqtlCaseData.py | 48 | ||||
| -rw-r--r-- | wqflask/base/webqtlConfig.py | 46 |
7 files changed, 766 insertions, 731 deletions
diff --git a/wqflask/base/GeneralObject.py b/wqflask/base/GeneralObject.py index 02a1ef06..ce8e60b8 100644 --- a/wqflask/base/GeneralObject.py +++ b/wqflask/base/GeneralObject.py @@ -25,44 +25,42 @@ # Last updated by GeneNetwork Core Team 2010/10/20 class GeneralObject: - """ - Base class to define an Object. - a = [Spam(1, 4), Spam(9, 3), Spam(4,6)] - a.sort(lambda x, y: cmp(x.eggs, y.eggs)) - """ + """ + Base class to define an Object. + a = [Spam(1, 4), Spam(9, 3), Spam(4,6)] + a.sort(key = lambda x: x.eggs) + """ - def __init__(self, *args, **kw): - self.contents = list(args) - for name, value in kw.items(): - setattr(self, name, value) - - def __setitem__(self, key, value): - setattr(self, key, value) - - def __getitem__(self, key): - return getattr(self, key) - - def __getattr__(self, key): - if key in self.__dict__.keys(): - return self.__dict__[key] - else: - return eval("self.__dict__.%s" % key) - - def __len__(self): - return len(self.__dict__) - 1 - - def __str__(self): - s = '' - for key in self.__dict__.keys(): - if key != 'contents': - s += '%s = %s\n' % (key,self.__dict__[key]) - return s - - def __repr__(self): - s = '' - for key in self.__dict__.keys(): - s += '%s = %s\n' % (key,self.__dict__[key]) - return s - - def __cmp__(self,other): - return len(self.__dict__.keys()).__cmp__(len(other.__dict__.keys()))
\ No newline at end of file + def __init__(self, *args, **kw): + self.contents = list(args) + for name, value in list(kw.items()): + setattr(self, name, value) + + def __setitem__(self, key, value): + setattr(self, key, value) + + def __getitem__(self, key): + return getattr(self, key) + + def __getattr__(self, key): + return eval("self.__dict__.%s" % key) + + def __len__(self): + return len(self.__dict__) - 1 + + def __str__(self): + s = '' + for key in list(self.__dict__.keys()): + if key != 'contents': + s += '%s = %s\n' % (key, self.__dict__[key]) + return s + + def __repr__(self): + s = '' + for key in list(self.__dict__.keys()): + s += '%s = %s\n' % (key, self.__dict__[key]) + return s + + def __eq__(self, other): + return (len(list(self.__dict__.keys())) + == len(list(other.__dict__.keys()))) diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index 1457ba8d..8906ab69 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -17,8 +17,28 @@ # at rwilliams@uthsc.edu and xzhou15@uthsc.edu # # This module is used by GeneNetwork project (www.genenetwork.org) - -from __future__ import absolute_import, print_function, division +from dataclasses import dataclass +from dataclasses import field +from dataclasses import InitVar +from typing import Optional, Dict +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 utility.db_tools import escape +from utility.db_tools import mescape +from utility.db_tools import create_in_clause +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 +from flask import Flask, g import os import math import string @@ -28,43 +48,25 @@ import codecs import json import requests import gzip -import cPickle as pickle +import pickle as pickle import itertools 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 import gen_geno_ob -from utility.tools import locate, locate_ignore_error, flat_files - -from wqflask.api import gen_menu - -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, GN2_BASE_URL -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) @@ -75,11 +77,11 @@ 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): - """Create a dictionary of samples where the value is set to Geno, -Publish or ProbeSet. E.g. +@dataclass +class DatasetType: + """Create a dictionary of samples where the value is set to Geno, + Publish or ProbeSet. E.g. {'AD-cases-controls-MyersGeno': 'Geno', 'AD-cases-controls-MyersPublish': 'Publish', @@ -90,20 +92,28 @@ Publish or ProbeSet. E.g. 'All Phenotypes': 'Publish', 'B139_K_1206_M': 'ProbeSet', 'B139_K_1206_R': 'ProbeSet' ... - + } """ - self.datasets = {} + redis_instance: InitVar[Redis] + datasets: Optional[Dict] = field(init=False, default_factory=dict) + data: Optional[Dict] = field(init=False) - data = Redis.get("dataset_structure") + def __post_init__(self, redis_instance): + self.redis_instance = redis_instance + data = 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 + 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]: + 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" @@ -112,84 +122,77 @@ Publish or ProbeSet. E.g. else: new_type = "ProbeSet" self.datasets[short_dataset_name] = new_type - except: + except Exception: # Do nothing pass - Redis.set("dataset_structure", json.dumps(self.datasets)) + self.redis_instance.set("dataset_structure", + json.dumps(self.datasets)) + self.data = data - # Set LOG_LEVEL_DEBUG=5 to see the following: - logger.debugf(5, "datasets",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 - def __call__(self, name): - if name not in self.datasets: - mrna_expr_query = """ - SELECT - ProbeSetFreeze.Id - FROM - ProbeSetFreeze - WHERE - ProbeSetFreeze.Name = "{0}" - """.format(name) + """ + sql_query_mapping = { + 'mrna_expr': ("SELECT ProbeSetFreeze.Id FROM " + "ProbeSetFreeze WHERE " + "ProbeSetFreeze.Name = \"%s\" "), + 'pheno': ("SELECT InfoFiles.GN_AccesionId " + "FROM InfoFiles, PublishFreeze, InbredSet " + "WHERE InbredSet.Name = '%s' AND " + "PublishFreeze.InbredSetId = InbredSet.Id AND " + "InfoFiles.InfoPageName = PublishFreeze.Name"), + 'other_pheno': ("SELECT PublishFreeze.Name " + "FROM PublishFreeze, InbredSet " + "WHERE InbredSet.Name = '%s' AND " + "PublishFreeze.InbredSetId = InbredSet.Id"), + 'geno': ("SELECT GenoFreeze.Id FROM GenoFreeze WHERE " + "GenoFreeze.Name = \"%s\" ") + } + + dataset_name_mapping = { + "mrna_expr": "ProbeSet", + "pheno": "Publish", + "other_pheno": "Publish", + "geno": "Geno", + } + + group_name = name + if t in ['pheno', 'other_pheno']: + group_name = name.replace("Publish", "") - results = g.db.execute(mrna_expr_query).fetchall() - if len(results): - self.datasets[name] = "ProbeSet" - Redis.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] + results = g.db.execute(sql_query_mapping[t] % group_name).fetchone() + if results: + self.datasets[name] = dataset_name_mapping[t] + self.redis_instance.set( + "dataset_structure", json.dumps(self.datasets)) + return True + return None - group_name = name.replace("Publish", "") + def __call__(self, 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 None if name has not been set + return self.datasets.get(name, None) - pheno_query = """SELECT InfoFiles.GN_AccesionId - FROM InfoFiles, PublishFreeze, InbredSet - WHERE InbredSet.Name = '{0}' AND - PublishFreeze.InbredSetId = InbredSet.Id AND - InfoFiles.InfoPageName = PublishFreeze.Name""".format(group_name) - - results = g.db.execute(pheno_query).fetchall() - if len(results): - self.datasets[name] = "Publish" - Redis.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - #ZS: For when there isn't an InfoFiles ID; not sure if this and the preceding query are both necessary - other_pheno_query = """SELECT PublishFreeze.Name - FROM PublishFreeze, InbredSet - WHERE InbredSet.Name = '{}' AND - PublishFreeze.InbredSetId = InbredSet.Id""".format(group_name) - - results = g.db.execute(other_pheno_query).fetchall() - if len(results): - self.datasets[name] = "Publish" - Redis.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - geno_query = """ - SELECT - GenoFreeze.Id - FROM - GenoFreeze - WHERE - GenoFreeze.Name = "{0}" - """.format(name) - - results = g.db.execute(geno_query).fetchall() - if len(results): - self.datasets[name] = "Geno" - Redis.set("dataset_structure", json.dumps(self.datasets)) - return self.datasets[name] - - #ZS: It shouldn't ever reach this - return None - else: - return self.datasets[name] # 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") @@ -205,38 +208,25 @@ 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 -def create_in_clause(items): - """Create an in clause for mysql""" - in_clause = ', '.join("'{}'".format(x) for x in mescape(*items)) - in_clause = '( {} )'.format(in_clause) - return in_clause - - -def mescape(*items): - """Multiple escape""" - escaped = [escape(str(item)) for item in items] - #logger.debug("escaped is:", escaped) - return escaped - - -class Markers(object): +class Markers: """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')) + 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: @@ -251,7 +241,8 @@ class Markers(object): for line in bimbam_fh: marker = {} marker['name'] = line.split(delimiter)[0].rstrip() - marker['Mb'] = float(line.split(delimiter)[1].rstrip())/1000000 + marker['Mb'] = float(line.split(delimiter)[ + 1].rstrip()) / 1000000 marker['chr'] = line.split(delimiter)[2].rstrip() markers.append(marker) @@ -266,12 +257,12 @@ class Markers(object): logger.debug("length of self.markers:", len(self.markers)) logger.debug("length of p_values:", len(p_values)) - if type(p_values) is list: + if isinstance(p_values, 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): + for marker, p_value in zip(self.markers, p_values): if not p_value: continue marker['p_value'] = float(p_value) @@ -280,38 +271,32 @@ 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: + elif isinstance(p_values, dict): filtered_markers = [] for marker in self.markers: - #logger.debug("marker[name]", marker['name']) - #logger.debug("p_values:", p_values) if marker['name'] in p_values: - #logger.debug("marker {} IS in p_values".format(i)) marker['p_value'] = p_values[marker['name']] if math.isnan(marker['p_value']) or (marker['p_value'] <= 0): marker['lod_score'] = 0 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: - #logger.debug("marker {} NOT in p_values".format(i)) - #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: splat = line.strip().split() - #logger.debug("splat:", splat) if len(specified_markers) > 0: if splat[1] in specified_markers: marker = {} @@ -327,14 +312,11 @@ class HumanMarkers(Markers): marker['Mb'] = float(splat[3]) / 1000000 self.markers.append(marker) - #logger.debug("markers is: ", pf(self.markers)) - - def add_pvalues(self, p_values): super(HumanMarkers, self).add_pvalues(p_values) -class DatasetGroup(object): +class DatasetGroup: """ Each group has multiple datasets; each species has multiple groups. @@ -342,12 +324,15 @@ class DatasetGroup(object): has multiple datasets associated with it. """ + def __init__(self, dataset, name=None): """This sets self.group and self.group_id""" 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" @@ -366,7 +351,8 @@ class DatasetGroup(object): 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 = ["GEMMA", "QTLReaper", "R/qtl"] elif mapping_id == "2": @@ -383,8 +369,8 @@ class DatasetGroup(object): def get_markers(self): def check_plink_gemma(): if flat_file_exists("mapping"): - MAPPING_PATH = flat_files("mapping")+"/" - if os.path.isfile(MAPPING_PATH+self.name+".bed"): + MAPPING_PATH = flat_files("mapping") + "/" + if os.path.isfile(MAPPING_PATH + self.name + ".bed"): return True return False @@ -410,6 +396,15 @@ class DatasetGroup(object): if maternal and paternal: self.parlist = [maternal, paternal] + def get_study_samplelists(self): + study_sample_file = locate_ignore_error(self.name + ".json", 'study_sample_lists') + try: + f = open(study_sample_file) + except: + return [] + study_samples = json.load(f) + return study_samples + def get_genofiles(self): jsonfile = "%s/%s.json" % (webqtlConfig.GENODIR, self.name) try: @@ -423,22 +418,23 @@ class DatasetGroup(object): result = None key = "samplelist:v3:" + self.name if USE_REDIS: - result = Redis.get(key) + result = r.get(key) if result is not None: self.samplelist = json.loads(result) else: logger.debug("Cache not hit") - genotype_fn = locate_ignore_error(self.name+".geno",'genotype') + genotype_fn = locate_ignore_error(self.name + ".geno", 'genotype') if genotype_fn: - self.samplelist = get_group_samplelists.get_samplelist("geno", genotype_fn) + self.samplelist = get_group_samplelists.get_samplelist( + "geno", genotype_fn) else: self.samplelist = None 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 = [] @@ -448,32 +444,28 @@ class DatasetGroup(object): 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() # reaper barfs on unicode filenames, so here we ensure it's a string if self.genofile: - 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')) + 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')) - - if use_reaper: - genotype_1 = reaper.Dataset() - genotype_1.read(full_filename) - else: - genotype_1 = gen_geno_ob.genotype(full_filename) + 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: @@ -484,27 +476,21 @@ 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 - and PublishFreeze.confidentiality < 1 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 GenoFreeze.confidentiality < 1) + and InbredSet.Name = '%s') UNION (SELECT Tissue.Name, ProbeSetFreeze.FullName,ProbeSetFreeze.Name FROM ProbeSetFreeze, ProbeFreeze, InbredSet, Tissue @@ -512,16 +498,15 @@ 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 - and ProbeSetFreeze.confidentiality < 1 ORDER BY Tissue.Name, ProbeSetFreeze.OrderList DESC) - ''' % (group_name, webqtlConfig.PUBLICTHRESH, - group_name, webqtlConfig.PUBLICTHRESH, - "'" + group_name + "'", webqtlConfig.PUBLICTHRESH)) + ''' % (group_name, + group_name, + "'" + group_name + "'")) sorted_results = sorted(the_results, key=lambda kv: kv[0]) - pheno_inserted = False #ZS: This is kind of awkward, but need to ensure Phenotypes show up before Genotypes in dropdown + # 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] @@ -529,13 +514,16 @@ def datasets(group_name, this_group = None): dataset_short = dataset_item[2] if tissue_name in ['#PublishFreeze', '#GenoFreeze']: if tissue_name == '#PublishFreeze' and (dataset_short == group_name + 'Publish'): - dataset_menu.insert(0, dict(tissue=None, datasets=[(dataset, dataset_short)])) + 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)])) + 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)])) + dataset_menu.append( + dict(tissue=None, datasets=[(dataset, dataset_short)])) else: tissue_already_exists = False for i, tissue_dict in enumerate(dataset_menu): @@ -544,15 +532,14 @@ def datasets(group_name, this_group = None): break if tissue_already_exists: - #logger.debug("dataset_menu:", dataset_menu[i]['datasets']) 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 @@ -560,14 +547,15 @@ def datasets(group_name, this_group = None): else: return dataset_menu -class DataSet(object): + +class DataSet: """ DataSet class defines a dataset in webqtl, can be either Microarray, Published phenotype, genotype, or user input dataset(temp) """ - 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 @@ -575,30 +563,34 @@ 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.accession_id = None self.setup() - if self.type == "Temp": #Need to supply group name as input if temp trait - self.group = DatasetGroup(self, name=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 - - # Delete this eventually - @property - def riset(): - Weve_Renamed_This_As_Group + def as_dict(self): + return { + 'name': self.name, + 'shortname': self.shortname, + 'fullname': self.fullname, + 'type': self.type, + 'data_scale': self.data_scale, + 'group': self.group.name, + 'accession_id': self.accession_id + } def get_accession_id(self): if self.type == "Publish": @@ -637,29 +629,26 @@ 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)) @@ -668,14 +657,77 @@ 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 chunk_dataset(self, dataset, n): + + results = {} + + query = """ + SELECT ProbeSetXRef.DataId,ProbeSet.Name + FROM ProbeSet, ProbeSetXRef, ProbeSetFreeze + WHERE ProbeSetFreeze.Name = '{}' AND + ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND + ProbeSetXRef.ProbeSetId = ProbeSet.Id + """.format(self.name) + + # should cache this + + traits_name_dict = dict(g.db.execute(query).fetchall()) + + for i in range(0, len(dataset), n): + matrix = list(dataset[i:i + n]) + trait_name = traits_name_dict[matrix[0][0]] + + my_values = [value for (trait_name, strain, value) in matrix] + results[trait_name] = my_values + return results + + def get_probeset_data(self, sample_list=None, trait_ids=None): + + # improvement of get trait data--->>> + if sample_list: + self.samplelist = sample_list + + else: + self.samplelist = self.group.samplelist + + if self.group.parlist != None and self.group.f1list != None: + if (self.group.parlist + self.group.f1list) in self.samplelist: + self.samplelist += self.group.parlist + self.group.f1list + + query = """ + SELECT Strain.Name, Strain.Id FROM Strain, Species + WHERE Strain.Name IN {} + and Strain.SpeciesId=Species.Id + and Species.name = '{}' + """.format(create_in_clause(self.samplelist), *mescape(self.group.species)) + results = dict(g.db.execute(query).fetchall()) + sample_ids = [results[item] for item in self.samplelist] + + sorted_samplelist = [strain_name for strain_name, strain_id in sorted( + results.items(), key=lambda item: item[1])] + + query = """SELECT * from ProbeSetData + where StrainID in {} + and id in (SELECT ProbeSetXRef.DataId + FROM (ProbeSet, ProbeSetXRef, ProbeSetFreeze) + WHERE ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id + and ProbeSetFreeze.Name = '{}' + and ProbeSet.Id = ProbeSetXRef.ProbeSetId)""".format(create_in_clause(sample_ids), self.name) + + query_results = list(g.db.execute(query).fetchall()) + data_results = self.chunk_dataset(query_results, len(sample_ids)) + self.samplelist = sorted_samplelist + self.trait_data = data_results + + def get_trait_data(self, sample_list=None): if sample_list: self.samplelist = sample_list @@ -692,7 +744,6 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre and Strain.SpeciesId=Species.Id and Species.name = '{}' """.format(create_in_clause(self.samplelist), *mescape(self.group.species)) - logger.sql(query) results = dict(g.db.execute(query).fetchall()) sample_ids = [results[item] for item in self.samplelist] @@ -713,7 +764,7 @@ AND (ProbeSetFreeze.Name = '%s' OR ProbeSetFreeze.FullName = '%s' OR ProbeSetFre else: query = "SELECT {}.Name,".format(escape(dataset_type)) data_start_pos = 1 - query += string.join(temp, ', ') + query += ', '.join(temp) query += ' FROM ({}, {}XRef, {}Freeze) '.format(*mescape(dataset_type, self.type, self.type)) @@ -731,7 +782,7 @@ 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 @@ -739,13 +790,9 @@ 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) - - logger.sql(query) results = g.db.execute(query).fetchall() - #logger.debug("query results:", results) trait_sample_data.append(results) trait_count = len(trait_sample_data[0]) @@ -759,25 +806,23 @@ 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' def setup(self): - - #logger.debug("IS A PHENOTYPEDATASET") - # Fields in the database table self.search_fields = ['Phenotype.Post_publication_description', - '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'] + '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', 'group_code', @@ -799,13 +844,13 @@ 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' @@ -823,7 +868,7 @@ class PhenotypeDataSet(DataSet): # (Urgently?) Need to write this pass - 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: @@ -831,9 +876,9 @@ 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, because no authorization features @@ -858,7 +903,7 @@ 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_location_repr = "N/A" @@ -869,7 +914,6 @@ class PhenotypeDataSet(DataSet): Geno.Name = '%s' and Geno.SpeciesId = Species.Id """ % (species, this_trait.locus) - logger.sql(query) result = g.db.execute(query).fetchone() if result: @@ -878,7 +922,8 @@ class PhenotypeDataSet(DataSet): LRS_Mb = result[1] this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % 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 = """ @@ -898,7 +943,6 @@ class PhenotypeDataSet(DataSet): Order BY Strain.Name """ - logger.sql(query) results = g.db.execute(query, (trait, self.id)).fetchall() return results @@ -945,12 +989,13 @@ class GenotypeDataSet(DataSet): this_trait.retrieveInfo() if this_trait.chr and this_trait.mb: - this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, float(this_trait.mb) ) + 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 @@ -964,7 +1009,6 @@ class GenotypeDataSet(DataSet): Order BY Strain.Name """ - logger.sql(query) results = g.db.execute(query, (webqtlDatabaseFunction.retrieve_species_id(self.group.name), trait, self.name)).fetchall() @@ -1014,14 +1058,14 @@ 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' @@ -1037,7 +1081,6 @@ class MrnaAssayDataSet(DataSet): ProbeSetFreeze.Name = "%s" ''' % escape(self.name) - def check_confidentiality(self): return geno_mrna_confidentiality(self) @@ -1055,37 +1098,37 @@ 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 = str( + str(this_trait.description).strip(codecs.BOM_UTF8), 'utf-8') + target_string = str( + 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 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'): + 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 template this_trait.description_display = description_display if this_trait.chr and this_trait.mb: - this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, float(this_trait.mb)) + this_trait.location_repr = 'Chr%s: %.6f' % ( + this_trait.chr, float(this_trait.mb)) - #Get mean expression value + # 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' """ % (escape(str(this_trait.dataset.id)), escape(this_trait.name))) - - #logger.debug("query is:", pf(query)) - logger.sql(query) result = g.db.execute(query).fetchone() mean = result[0] if result else 0 @@ -1093,11 +1136,11 @@ 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_location_repr = 'N/A' - #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 @@ -1105,24 +1148,28 @@ class MrnaAssayDataSet(DataSet): Geno.Name = '{}' and Geno.SpeciesId = Species.Id """.format(species, this_trait.locus) - logger.sql(query) result = g.db.execute(query).fetchone() if result: lrs_chr, lrs_mb = result this_trait.LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_location_repr = 'Chr%s: %.6f' % (lrs_chr, float(lrs_mb)) + this_trait.LRS_location_repr = 'Chr%s: %.6f' % ( + lrs_chr, float(lrs_mb)) 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) + (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 @@ -1132,9 +1179,7 @@ class MrnaAssayDataSet(DataSet): Order BY Strain.Name """ % (escape(trait), escape(self.name)) - logger.sql(query) results = g.db.execute(query).fetchall() - #logger.debug("RETRIEVED RESULTS HERE:", results) return results def retrieve_genes(self, column_name): @@ -1144,7 +1189,6 @@ class MrnaAssayDataSet(DataSet): where ProbeSetXRef.ProbeSetFreezeId = %s and ProbeSetXRef.ProbeSetId=ProbeSet.Id; """ % (column_name, escape(str(self.id))) - logger.sql(query) results = g.db.execute(query).fetchall() return dict(results) @@ -1173,48 +1217,21 @@ class TempDataSet(DataSet): self.shortname = 'Temp' - @staticmethod - def handle_pca(desc): - if 'PCA' in desc: - # Todo: Modernize below lines - desc = desc[desc.rindex(':')+1:].strip() - else: - desc = desc[:desc.index('entered')].strip() - return desc - - def get_desc(self): - query = 'SELECT description FROM Temp WHERE Name=%s' % self.name - logger.sql(query) - g.db.execute(query) - desc = g.db.fetchone()[0] - desc = self.handle_pca(desc) - return desc - - def retrieve_sample_data(self, trait): - query = """ - SELECT - Strain.Name, TempData.value, TempData.SE, TempData.NStrain, TempData.Id - FROM - TempData, Temp, Strain - WHERE - TempData.StrainId = Strain.Id AND - TempData.Id = Temp.DataId AND - Temp.name = '%s' - Order BY - Strain.Name - """ % escape(trait.name) - - logger.sql(query) - results = g.db.execute(query).fetchall() - - def geno_mrna_confidentiality(ob): dataset_table = ob.type + "Freeze" - #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) - logger.sql(query) + AuthorisedUsers FROM %s WHERE Name = "%s"''' % (dataset_table, ob.name) + result = g.db.execute(query) + + (dataset_id, + name, + full_name, + confidential, + authorized_users) = result.fetchall()[0] + + if confidential: + return True result = g.db.execute(query) (dataset_id, diff --git a/wqflask/base/mrna_assay_tissue_data.py b/wqflask/base/mrna_assay_tissue_data.py index 6fec5dcd..8f8e2b0a 100644 --- a/wqflask/base/mrna_assay_tissue_data.py +++ b/wqflask/base/mrna_assay_tissue_data.py @@ -1,5 +1,3 @@ -from __future__ import absolute_import, print_function, division - import collections from flask import g @@ -7,14 +5,15 @@ from flask import g from utility import db_tools from utility import Bunch -from MySQLdb import escape_string as escape +from utility.db_tools import escape +from gn3.db_utils import database_connector -from pprint import pformat as pf from utility.logger import getLogger -logger = getLogger(__name__ ) +logger = getLogger(__name__) + -class MrnaAssayTissueData(object): +class MrnaAssayTissueData: def __init__(self, gene_symbols=None): self.gene_symbols = gene_symbols @@ -23,7 +22,7 @@ class MrnaAssayTissueData(object): self.data = collections.defaultdict(Bunch) - query = '''select t.Symbol, t.GeneId, t.DataId, t.Chr, t.Mb, t.description, t.Probe_Target_Description + query = '''select t.Symbol, t.GeneId, t.DataId, t.Chr, t.Mb, t.description, t.Probe_Target_Description from ( select Symbol, max(Mean) as maxmean from TissueProbeSetXRef @@ -34,29 +33,31 @@ class MrnaAssayTissueData(object): # Due to the limit size of TissueProbeSetFreezeId table in DB, # performance of inner join is acceptable.MrnaAssayTissueData(gene_symbols=symbol_list) if len(gene_symbols) == 0: - query += '''Symbol!='' and Symbol Is Not Null group by Symbol) + query += '''Symbol!='' and Symbol Is Not Null group by Symbol) as x inner join TissueProbeSetXRef as t on t.Symbol = x.Symbol and t.Mean = x.maxmean; ''' else: in_clause = db_tools.create_in_clause(gene_symbols) - #ZS: This was in the query, not sure why: http://docs.python.org/2/library/string.html?highlight=lower#string.lower + # ZS: This was in the query, not sure why: http://docs.python.org/2/library/string.html?highlight=lower#string.lower query += ''' Symbol in {} group by Symbol) as x inner join TissueProbeSetXRef as t on t.Symbol = x.Symbol and t.Mean = x.maxmean; '''.format(in_clause) - results = g.db.execute(query).fetchall() - lower_symbols = [] + # lower_symbols = [] + lower_symbols = {} for gene_symbol in gene_symbols: + # lower_symbols[gene_symbol.lower()] = True if gene_symbol != None: - lower_symbols.append(gene_symbol.lower()) - + lower_symbols[gene_symbol.lower()] = True + results = list(g.db.execute(query).fetchall()) for result in results: symbol = result[0] - if symbol.lower() in lower_symbols: + if symbol is not None and lower_symbols.get(symbol.lower()): + symbol = symbol.lower() self.data[symbol].gene_id = result.GeneId @@ -67,16 +68,16 @@ class MrnaAssayTissueData(object): self.data[symbol].probe_target_description = result.Probe_Target_Description ########################################################################### - #Input: cursor, symbolList (list), dataIdDict(Dict) - #output: symbolValuepairDict (dictionary):one dictionary of Symbol and Value Pair, + # Input: cursor, symbolList (list), dataIdDict(Dict) + # output: symbolValuepairDict (dictionary):one dictionary of Symbol and Value Pair, # key is symbol, value is one list of expression values of one probeSet; - #function: get one dictionary whose key is gene symbol and value is tissue expression data (list type). - #Attention! All keys are lower case! + # function: get one dictionary whose key is gene symbol and value is tissue expression data (list type). + # Attention! All keys are lower case! ########################################################################### def get_symbol_values_pairs(self): id_list = [self.data[symbol].data_id for symbol in self.data] - + symbol_values_dict = {} if len(id_list) > 0: @@ -85,11 +86,13 @@ class MrnaAssayTissueData(object): WHERE TissueProbeSetData.Id IN {} and TissueProbeSetXRef.DataId = TissueProbeSetData.Id""".format(db_tools.create_in_clause(id_list)) + results = g.db.execute(query).fetchall() for result in results: if result.Symbol.lower() not in symbol_values_dict: symbol_values_dict[result.Symbol.lower()] = [result.value] else: - symbol_values_dict[result.Symbol.lower()].append(result.value) + symbol_values_dict[result.Symbol.lower()].append( + result.value) - return symbol_values_dict
\ No newline at end of file + return symbol_values_dict diff --git a/wqflask/base/species.py b/wqflask/base/species.py index 6d99af65..f303aabb 100644 --- a/wqflask/base/species.py +++ b/wqflask/base/species.py @@ -1,62 +1,66 @@ -from __future__ import absolute_import, print_function, division +from collections import OrderedDict +from dataclasses import dataclass +from dataclasses import InitVar +from typing import Optional, Dict +from flask import g -import collections -from flask import Flask, g +@dataclass +class TheSpecies: + """Data related to species.""" + dataset: Optional[Dict] = None + species_name: Optional[str] = None -#from MySQLdb import escape_string as escape - -from utility import Bunch - -from pprint import pformat as pf - -from utility.logger import getLogger -logger = getLogger(__name__ ) - -class TheSpecies(object): - def __init__(self, dataset=None, species_name=None): - if species_name != None: - self.name = species_name + def __post_init__(self): + if self.species_name is not None: + self.name = self.species_name self.chromosomes = Chromosomes(species=self.name) else: - self.dataset = dataset self.chromosomes = Chromosomes(dataset=self.dataset) -class IndChromosome(object): - def __init__(self, name, length): - self.name = name - self.length = length + +@dataclass +class IndChromosome: + """Data related to IndChromosome""" + name: str + length: int @property def mb_length(self): - """Chromosome length in megabases""" + """Chromosome length in mega-bases""" return self.length / 1000000 -class Chromosomes(object): - def __init__(self, dataset=None, species=None): - self.chromosomes = collections.OrderedDict() - if species != None: - query = """ - Select - Chr_Length.Name, Chr_Length.OrderId, Length from Chr_Length, Species - where - Chr_Length.SpeciesId = Species.SpeciesId AND - Species.Name = '%s' - Order by OrderId - """ % species.capitalize() - else: + +@dataclass +class Chromosomes: + """Data related to a chromosome""" + dataset: InitVar[Dict] = None + species: Optional[str] = None + + def __post_init__(self, dataset): + if self.species is None: self.dataset = dataset - 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 - """ % self.dataset.group.name - logger.sql(query) + @property + def chromosomes(self): + """Lazily fetch the chromosomes""" + chromosomes = OrderedDict() + if self.species is not None: + query = ( + "SELECT Chr_Length.Name, Chr_Length.OrderId, Length " + "FROM Chr_Length, Species WHERE " + "Chr_Length.SpeciesId = Species.SpeciesId AND " + "Species.Name = " + "'%s' ORDER BY OrderId" % self.species.capitalize()) + else: + 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" % self.dataset.group.name) results = g.db.execute(query).fetchall() - for item in results: - self.chromosomes[item.OrderId] = IndChromosome(item.Name, item.Length)
\ No newline at end of file + chromosomes[item.OrderId] = IndChromosome( + item.Name, item.Length) + return chromosomes diff --git a/wqflask/base/trait.py b/wqflask/base/trait.py index e454c593..f0749858 100644 --- a/wqflask/base/trait.py +++ b/wqflask/base/trait.py @@ -1,33 +1,59 @@ -from __future__ import absolute_import, division, print_function - -import string -import resource -import codecs import requests - -import redis -Redis = redis.StrictRedis() +import simplejson as json +from wqflask import app from base import webqtlConfig from base.webqtlCaseData import webqtlCaseData from base.data_set import create_dataset -from db import webqtlDatabaseFunction -from utility import webqtlUtil from utility import hmac -from utility.tools import GN2_BASE_URL +from utility.authentication_tools import check_resource_availability +from utility.tools import GN2_BASE_URL, GN_PROXY_URL +from utility.redis_tools import get_redis_conn, get_resource_id -from wqflask import app - -import simplejson as json -from MySQLdb import escape_string as escape -from pprint import pformat as pf +from utility.db_tools import escape -from flask import Flask, g, request, url_for +from flask import g, request, url_for from utility.logger import getLogger -logger = getLogger(__name__ ) -class GeneralTrait(object): +logger = getLogger(__name__) + +Redis = get_redis_conn() + + +def create_trait(**kw): + assert bool(kw.get('dataset')) != bool( + kw.get('dataset_name')), "Needs dataset ob. or name" + + assert bool(kw.get('name')), "Needs trait name" + + if bool(kw.get('dataset')): + dataset = kw.get('dataset') + else: + if kw.get('dataset_name') != "Temp": + dataset = create_dataset(kw.get('dataset_name')) + else: + dataset = create_dataset("Temp", group_name=kw.get('group_name')) + + if dataset.type == 'Publish': + permissions = check_resource_availability( + dataset, kw.get('name')) + else: + permissions = check_resource_availability(dataset) + + if permissions['data'] != "no-access": + the_trait = GeneralTrait(**kw) + if the_trait.dataset.type != "Temp": + the_trait = retrieve_trait_info( + the_trait, + the_trait.dataset, + get_qtl_info=kw.get('get_qtl_info')) + return the_trait + else: + return None + + +class GeneralTrait: """ Trait class defines a trait in webqtl, can be either Microarray, Published phenotype, genotype, or user input trait @@ -36,12 +62,17 @@ class GeneralTrait(object): def __init__(self, get_qtl_info=False, get_sample_info=True, **kw): # xor assertion - assert bool(kw.get('dataset')) != bool(kw.get('dataset_name')), "Needs dataset ob. or name"; - self.name = kw.get('name') # Trait ID, ProbeSet ID, Published ID, etc. + assert bool(kw.get('dataset')) != bool( + kw.get('dataset_name')), "Needs dataset ob. or name" + # Trait ID, ProbeSet ID, Published ID, etc. + self.name = kw.get('name') if kw.get('dataset_name'): if kw.get('dataset_name') == "Temp": temp_group = self.name.split("_")[2] - self.dataset = create_dataset(dataset_name = "Temp", dataset_type = "Temp", group_name = temp_group) + self.dataset = create_dataset( + dataset_name="Temp", + dataset_type="Temp", + group_name=temp_group) else: self.dataset = create_dataset(kw.get('dataset_name')) else: @@ -49,8 +80,10 @@ class GeneralTrait(object): self.cellid = kw.get('cellid') self.identification = kw.get('identification', 'un-named trait') self.haveinfo = kw.get('haveinfo', False) - self.sequence = kw.get('sequence') # Blat sequence, available for ProbeSet + # Blat sequence, available for ProbeSet + self.sequence = kw.get('sequence') self.data = kw.get('data', {}) + self.view = True # Sets defaults self.locus = None @@ -74,11 +107,10 @@ class GeneralTrait(object): elif len(name2) == 3: self.dataset, self.name, self.cellid = name2 - # Todo: These two lines are necessary most of the time, but perhaps not all of the time - # So we could add a simple if statement to short-circuit this if necessary - if self.dataset.type != "Temp": - self = retrieve_trait_info(self, self.dataset, get_qtl_info=get_qtl_info) - if get_sample_info != False: + # Todo: These two lines are necessary most of the time, but + # perhaps not all of the time So we could add a simple if + # statement to short-circuit this if necessary + if get_sample_info is not False: self = retrieve_sample_data(self, self.dataset) def export_informative(self, include_variance=0): @@ -91,18 +123,18 @@ class GeneralTrait(object): vals = [] the_vars = [] sample_aliases = [] - for sample_name, sample_data in self.data.items(): - if sample_data.value != None: - if not include_variance or sample_data.variance != None: + for sample_name, sample_data in list(self.data.items()): + if sample_data.value is not None: + if not include_variance or sample_data.variance is not None: samples.append(sample_name) vals.append(sample_data.value) the_vars.append(sample_data.variance) sample_aliases.append(sample_data.name2) - return samples, vals, the_vars, sample_aliases + return samples, vals, the_vars, sample_aliases @property def description_fmt(self): - '''Return a text formated description''' + """Return a text formated description""" if self.dataset.type == 'ProbeSet': if self.description: formatted = self.description @@ -117,31 +149,38 @@ class GeneralTrait(object): formatted = self.post_publication_description else: formatted = "Not available" + if isinstance(formatted, bytes): + formatted = formatted.decode("utf-8") return formatted @property def alias_fmt(self): - '''Return a text formatted alias''' + """Return a text formatted alias""" alias = 'Not available' - if self.alias: - alias = string.replace(self.alias, ";", " ") - alias = string.join(string.split(alias), ", ") + if getattr(self, "alias", None): + alias = self.alias.replace(";", " ") + alias = ", ".join(alias.split()) return alias @property def wikidata_alias_fmt(self): - '''Return a text formatted alias''' + """Return a text formatted alias""" alias = 'Not available' if self.symbol: - human_response = requests.get(GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.upper()) - mouse_response = requests.get(GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.capitalize()) - other_response = requests.get(GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.lower()) + human_response = requests.get( + GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.upper()) + mouse_response = requests.get( + GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.capitalize()) + other_response = requests.get( + GN2_BASE_URL + "gn3/gene/aliases/" + self.symbol.lower()) if human_response and mouse_response and other_response: - alias_list = json.loads(human_response.content) + json.loads(mouse_response.content) + json.loads(other_response.content) + alias_list = json.loads(human_response.content) + json.loads( + mouse_response.content) + \ + json.loads(other_response.content) filtered_aliases = [] seen = set() @@ -155,33 +194,34 @@ class GeneralTrait(object): return alias - @property def location_fmt(self): - '''Return a text formatted location + """Return a text formatted location - While we're at it we set self.location in case we need it later (do we?) + While we're at it we set self.location in case we need it + later (do we?) - ''' + """ if self.chr and self.mb: - self.location = 'Chr %s @ %s Mb' % (self.chr,self.mb) + self.location = 'Chr %s @ %s Mb' % (self.chr, self.mb) elif self.chr: self.location = 'Chr %s @ Unknown position' % (self.chr) else: self.location = 'Not available' fmt = self.location - ##XZ: deal with direction + # XZ: deal with direction if self.strand_probe == '+': fmt += (' on the plus strand ') elif self.strand_probe == '-': fmt += (' on the minus strand ') return fmt - + + def retrieve_sample_data(trait, dataset, samplelist=None): - if samplelist == None: + if samplelist is None: samplelist = [] if dataset.type == "Temp": @@ -197,55 +237,71 @@ def retrieve_sample_data(trait, dataset, samplelist=None): all_samples_ordered = dataset.group.all_samples_ordered() for i, item in enumerate(results): try: - trait.data[all_samples_ordered[i]] = webqtlCaseData(all_samples_ordered[i], float(item)) + trait.data[all_samples_ordered[i]] = webqtlCaseData( + all_samples_ordered[i], float(item)) except: pass else: for item in results: name, value, variance, num_cases, name2 = item if not samplelist or (samplelist and name in samplelist): - trait.data[name] = webqtlCaseData(*item) #name, value, variance, num_cases) + # name, value, variance, num_cases) + trait.data[name] = webqtlCaseData(*item) return trait + @app.route("/trait/get_sample_data") def get_sample_data(): params = request.args trait = params['trait'] dataset = params['dataset'] - trait_ob = GeneralTrait(name=trait, dataset_name=dataset) - - trait_dict = {} - trait_dict['name'] = trait - trait_dict['db'] = dataset - trait_dict['type'] = trait_ob.dataset.type - trait_dict['group'] = trait_ob.dataset.group.name - trait_dict['tissue'] = trait_ob.dataset.tissue - trait_dict['species'] = trait_ob.dataset.group.species - trait_dict['url'] = url_for('show_trait_page', trait_id = trait, dataset = dataset) - trait_dict['description'] = trait_ob.description_display - if trait_ob.dataset.type == "ProbeSet": - trait_dict['symbol'] = trait_ob.symbol - trait_dict['location'] = trait_ob.location_repr - elif trait_ob.dataset.type == "Publish": - if trait_ob.pubmed_id: - trait_dict['pubmed_link'] = trait_ob.pubmed_link - trait_dict['pubmed_text'] = trait_ob.pubmed_text - - return json.dumps([trait_dict, {key: value.value for key, value in trait_ob.data.iteritems() }]) - -def jsonable(trait): + trait_ob = create_trait(name=trait, dataset_name=dataset) + if trait_ob: + trait_dict = {} + trait_dict['name'] = trait + trait_dict['db'] = dataset + trait_dict['type'] = trait_ob.dataset.type + trait_dict['group'] = trait_ob.dataset.group.name + trait_dict['tissue'] = trait_ob.dataset.tissue + trait_dict['species'] = trait_ob.dataset.group.species + trait_dict['url'] = url_for( + 'show_trait_page', trait_id=trait, dataset=dataset) + if trait_ob.dataset.type == "ProbeSet": + trait_dict['symbol'] = trait_ob.symbol + trait_dict['location'] = trait_ob.location_repr + trait_dict['description'] = trait_ob.description_display + elif trait_ob.dataset.type == "Publish": + trait_dict['description'] = trait_ob.description_display + if trait_ob.pubmed_id: + trait_dict['pubmed_link'] = trait_ob.pubmed_link + trait_dict['pubmed_text'] = trait_ob.pubmed_text + else: + trait_dict['location'] = trait_ob.location_repr + + return json.dumps([trait_dict, {key: value.value for + key, value in list( + trait_ob.data.items())}]) + else: + return None + + +def jsonable(trait, dataset=None): """Return a dict suitable for using as json Actual turning into json doesn't happen here though""" - dataset = create_dataset(dataset_name = trait.dataset.name, dataset_type = trait.dataset.type, group_name = trait.dataset.group.name) - + if not dataset: + dataset = create_dataset(dataset_name=trait.dataset.name, + dataset_type=trait.dataset.type, + group_name=trait.dataset.group.name) + if dataset.type == "ProbeSet": return dict(name=trait.name, + view=trait.view, symbol=trait.symbol, dataset=dataset.name, - dataset_name = dataset.shortname, + dataset_name=dataset.shortname, description=trait.description_display, mean=trait.mean, location=trait.location_repr, @@ -256,181 +312,142 @@ def jsonable(trait): elif dataset.type == "Publish": if trait.pubmed_id: return dict(name=trait.name, + view=trait.view, dataset=dataset.name, - dataset_name = dataset.shortname, + dataset_name=dataset.shortname, description=trait.description_display, abbreviation=trait.abbreviation, authors=trait.authors, + pubmed_id=trait.pubmed_id, pubmed_text=trait.pubmed_text, pubmed_link=trait.pubmed_link, + mean=trait.mean, lrs_score=trait.LRS_score_repr, lrs_location=trait.LRS_location_repr, additive=trait.additive ) else: return dict(name=trait.name, + view=trait.view, dataset=dataset.name, - dataset_name = dataset.shortname, + dataset_name=dataset.shortname, description=trait.description_display, abbreviation=trait.abbreviation, authors=trait.authors, pubmed_text=trait.pubmed_text, + mean=trait.mean, lrs_score=trait.LRS_score_repr, lrs_location=trait.LRS_location_repr, additive=trait.additive ) elif dataset.type == "Geno": return dict(name=trait.name, + view=trait.view, dataset=dataset.name, - dataset_name = dataset.shortname, + dataset_name=dataset.shortname, location=trait.location_repr ) + elif dataset.name == "Temp": + return dict(name=trait.name, + view=trait.view, + dataset="Temp", + dataset_name="Temp") else: return dict() -def jsonable_table_row(trait, dataset_name, index): - """Return a list suitable for json and intended to be displayed in a table - - Actual turning into json doesn't happen here though""" - - dataset = create_dataset(dataset_name) - - if dataset.type == "ProbeSet": - if trait.mean == "": - mean = "N/A" - else: - mean = "%.3f" % round(float(trait.mean), 2) - if trait.additive == "": - additive = "N/A" - else: - additive = "%.3f" % round(float(trait.additive), 2) - return ['<input type="checkbox" name="searchResult" class="checkbox trait_checkbox" value="' + hmac.data_hmac('{}:{}'.format(str(trait.name), dataset.name)) + '">', - index, - '<a href="/show_trait?trait_id='+str(trait.name)+'&dataset='+dataset.name+'">'+str(trait.name)+'</a>', - trait.symbol, - trait.description_display, - trait.location_repr, - mean, - trait.LRS_score_repr, - trait.LRS_location_repr, - additive] - elif dataset.type == "Publish": - if trait.additive == "": - additive = "N/A" - else: - additive = "%.2f" % round(float(trait.additive), 2) - if trait.pubmed_id: - return ['<input type="checkbox" name="searchResult" class="checkbox trait_checkbox" value="' + hmac.data_hmac('{}:{}'.format(str(trait.name), dataset.name)) + '">', - index, - '<a href="/show_trait?trait_id='+str(trait.name)+'&dataset='+dataset.name+'">'+str(trait.name)+'</a>', - trait.description_display, - trait.authors, - '<a href="' + trait.pubmed_link + '">' + trait.pubmed_text + '</href>', - trait.LRS_score_repr, - trait.LRS_location_repr, - additive] - else: - return ['<input type="checkbox" name="searchResult" class="checkbox trait_checkbox" value="' + hmac.data_hmac('{}:{}'.format(str(trait.name), dataset.name)) + '">', - index, - '<a href="/show_trait?trait_id='+str(trait.name)+'&dataset='+dataset.name+'">'+str(trait.name)+'</a>', - trait.description_display, - trait.authors, - trait.pubmed_text, - trait.LRS_score_repr, - trait.LRS_location_repr, - additive] - elif dataset.type == "Geno": - return ['<input type="checkbox" name="searchResult" class="checkbox trait_checkbox" value="' + hmac.data_hmac('{}:{}'.format(str(trait.name), dataset.name)) + '">', - index, - '<a href="/show_trait?trait_id='+str(trait.name)+'&dataset='+dataset.name+'">'+str(trait.name)+'</a>', - trait.location_repr] - else: - return dict() def retrieve_trait_info(trait, dataset, get_qtl_info=False): assert dataset, "Dataset doesn't exist" - + + resource_id = get_resource_id(dataset, trait.name) if dataset.type == 'Publish': - query = """ - SELECT - PublishXRef.Id, InbredSet.InbredSetCode, Publication.PubMed_ID, - Phenotype.Pre_publication_description, Phenotype.Post_publication_description, Phenotype.Original_description, - Phenotype.Pre_publication_abbreviation, Phenotype.Post_publication_abbreviation, PublishXRef.mean, - Phenotype.Lab_code, Phenotype.Submitter, Phenotype.Owner, Phenotype.Authorized_Users, - Publication.Authors, Publication.Title, Publication.Abstract, - Publication.Journal, Publication.Volume, Publication.Pages, - Publication.Month, Publication.Year, PublishXRef.Sequence, - Phenotype.Units, PublishXRef.comments - FROM - PublishXRef, Publication, Phenotype, PublishFreeze, InbredSet - WHERE - PublishXRef.Id = %s AND - Phenotype.Id = PublishXRef.PhenotypeId AND - Publication.Id = PublishXRef.PublicationId AND - PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND - PublishXRef.InbredSetId = InbredSet.Id AND - PublishFreeze.Id = %s - """ % (trait.name, dataset.id) + the_url = GN_PROXY_URL + "run-action?resource={}&user={}&branch=data&action=view".format( + resource_id, g.user_session.user_id) + else: + the_url = GN_PROXY_URL + "run-action?resource={}&user={}&branch=data&action=view&trait={}".format( + resource_id, g.user_session.user_id, trait.name) - logger.sql(query) - trait_info = g.db.execute(query).fetchone() - - - #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 dataset.type == 'ProbeSet': - display_fields_string = ', ProbeSet.'.join(dataset.display_fields) - display_fields_string = 'ProbeSet.' + display_fields_string - query = """ - SELECT %s - FROM ProbeSet, ProbeSetFreeze, ProbeSetXRef - WHERE - ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND - ProbeSetXRef.ProbeSetId = ProbeSet.Id AND - ProbeSetFreeze.Name = '%s' AND - ProbeSet.Name = '%s' - """ % (escape(display_fields_string), - escape(dataset.name), - escape(str(trait.name))) - logger.sql(query) - trait_info = g.db.execute(query).fetchone() - #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 dataset.type == 'Geno': - display_fields_string = string.join(dataset.display_fields,',Geno.') - display_fields_string = 'Geno.' + display_fields_string - query = """ - SELECT %s - FROM Geno, GenoFreeze, GenoXRef - WHERE - GenoXRef.GenoFreezeId = GenoFreeze.Id AND - GenoXRef.GenoId = Geno.Id AND - GenoFreeze.Name = '%s' AND - Geno.Name = '%s' - """ % (escape(display_fields_string), - escape(dataset.name), - escape(trait.name)) - logger.sql(query) - trait_info = g.db.execute(query).fetchone() - else: #Temp type - query = """SELECT %s FROM %s WHERE Name = %s""" - logger.sql(query) - trait_info = g.db.execute(query, - (string.join(dataset.display_fields,','), - dataset.type, trait.name)).fetchone() + try: + response = requests.get(the_url).content + trait_info = json.loads(response) + except: # ZS: I'm assuming the trait is viewable if the try fails for some reason; it should never reach this point unless the user has privileges, since that's dealt with in create_trait + if dataset.type == 'Publish': + query = """ + SELECT + PublishXRef.Id, InbredSet.InbredSetCode, Publication.PubMed_ID, + CAST(Phenotype.Pre_publication_description AS BINARY), + CAST(Phenotype.Post_publication_description AS BINARY), + CAST(Phenotype.Original_description AS BINARY), + CAST(Phenotype.Pre_publication_abbreviation AS BINARY), + CAST(Phenotype.Post_publication_abbreviation AS BINARY), PublishXRef.mean, + Phenotype.Lab_code, Phenotype.Submitter, Phenotype.Owner, Phenotype.Authorized_Users, + CAST(Publication.Authors AS BINARY), CAST(Publication.Title AS BINARY), CAST(Publication.Abstract AS BINARY), + CAST(Publication.Journal AS BINARY), Publication.Volume, Publication.Pages, + Publication.Month, Publication.Year, PublishXRef.Sequence, + Phenotype.Units, PublishXRef.comments + FROM + PublishXRef, Publication, Phenotype, PublishFreeze, InbredSet + WHERE + PublishXRef.Id = %s AND + Phenotype.Id = PublishXRef.PhenotypeId AND + Publication.Id = PublishXRef.PublicationId AND + PublishXRef.InbredSetId = PublishFreeze.InbredSetId AND + PublishXRef.InbredSetId = InbredSet.Id AND + PublishFreeze.Id = %s + """ % (trait.name, dataset.id) + + logger.sql(query) + trait_info = g.db.execute(query).fetchone() + + # 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 dataset.type == 'ProbeSet': + display_fields_string = ', ProbeSet.'.join(dataset.display_fields) + display_fields_string = 'ProbeSet.' + display_fields_string + query = """ + SELECT %s + FROM ProbeSet, ProbeSetFreeze, ProbeSetXRef + WHERE + ProbeSetXRef.ProbeSetFreezeId = ProbeSetFreeze.Id AND + ProbeSetXRef.ProbeSetId = ProbeSet.Id AND + ProbeSetFreeze.Name = '%s' AND + ProbeSet.Name = '%s' + """ % (escape(display_fields_string), + escape(dataset.name), + escape(str(trait.name))) + logger.sql(query) + trait_info = g.db.execute(query).fetchone() + # 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 dataset.type == 'Geno': + display_fields_string = ',Geno.'.join(dataset.display_fields) + display_fields_string = 'Geno.' + display_fields_string + query = """ + SELECT %s + FROM Geno, GenoFreeze, GenoXRef + WHERE + GenoXRef.GenoFreezeId = GenoFreeze.Id AND + GenoXRef.GenoId = Geno.Id AND + GenoFreeze.Name = '%s' AND + Geno.Name = '%s' + """ % (escape(display_fields_string), + escape(dataset.name), + escape(trait.name)) + logger.sql(query) + trait_info = g.db.execute(query).fetchone() + else: # Temp type + query = """SELECT %s FROM %s WHERE Name = %s""" + logger.sql(query) + trait_info = g.db.execute(query, + ','.join(dataset.display_fields), + dataset.type, trait.name).fetchone() if trait_info: trait.haveinfo = True - - #XZ: assign SQL query result to trait attributes. for i, field in enumerate(dataset.display_fields): holder = trait_info[i] - # if isinstance(trait_info[i], basestring): - # logger.debug("HOLDER:", holder) - # logger.debug("HOLDER2:", holder.decode(encoding='latin1')) - # holder = unicode(trait_info[i], "utf-8", "ignore") - if isinstance(trait_info[i], basestring): - holder = holder.encode('latin1') + if isinstance(holder, bytes): + holder = holder.decode("utf-8", errors="ignore") setattr(trait, field, holder) if dataset.type == 'Publish': @@ -443,29 +460,18 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False): description = 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 trait.confidential: + # 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 + trait.description_display = "N/A" + if not trait.pubmed_id: trait.abbreviation = trait.pre_publication_abbreviation - trait.description_display = trait.pre_publication_description - - #if not webqtlUtil.hasAccessToConfidentialPhenotypeTrait( - # privilege=self.dataset.privilege, - # userName=self.dataset.userName, - # authorized_users=self.authorized_users): - # - # description = self.pre_publication_description + if trait.pre_publication_description: + trait.description_display = trait.pre_publication_description else: trait.abbreviation = trait.post_publication_abbreviation if description: trait.description_display = description.strip() - else: - trait.description_display = "" - - trait.abbreviation = unicode(str(trait.abbreviation).strip(codecs.BOM_UTF8), 'utf-8', errors="replace") - trait.description_display = unicode(str(trait.description_display).strip(codecs.BOM_UTF8), 'utf-8', errors="replace") - trait.authors = unicode(str(trait.authors).strip(codecs.BOM_UTF8), 'utf-8', errors="replace") if not trait.year.isdigit(): trait.pubmed_text = "N/A" @@ -476,16 +482,17 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False): trait.pubmed_link = webqtlConfig.PUBMEDLINK_URL % trait.pubmed_id if dataset.type == 'ProbeSet' and dataset.group: - description_string = unicode(str(trait.description).strip(codecs.BOM_UTF8), 'utf-8') - target_string = unicode(str(trait.probe_target_description).strip(codecs.BOM_UTF8), 'utf-8') + description_string = trait.description + target_string = trait.probe_target_description - if len(description_string) > 1 and description_string != 'None': + if str(description_string or "") != "" and description_string != 'None': description_display = description_string else: description_display = trait.symbol - if (len(description_display) > 1 and description_display != 'N/A' and - len(target_string) > 1 and target_string != 'None'): + if (str(description_display or "") != "" + and description_display != 'N/A' + and str(target_string or "") != "" and target_string != 'None'): description_display = description_display + '; ' + target_string.strip() # Save it for the jinja2 template @@ -493,15 +500,17 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False): trait.location_repr = 'N/A' if trait.chr and trait.mb: - trait.location_repr = 'Chr%s: %.6f' % (trait.chr, float(trait.mb)) + trait.location_repr = 'Chr%s: %.6f' % ( + trait.chr, float(trait.mb)) elif dataset.type == "Geno": trait.location_repr = 'N/A' if trait.chr and trait.mb: - trait.location_repr = 'Chr%s: %.6f' % (trait.chr, float(trait.mb)) + trait.location_repr = 'Chr%s: %.6f' % ( + trait.chr, float(trait.mb)) if get_qtl_info: - #LRS and its location + # LRS and its location trait.LRS_score_repr = "N/A" trait.LRS_location_repr = "N/A" trait.locus = trait.locus_chr = trait.locus_mb = trait.lrs = trait.pvalue = trait.additive = "" @@ -571,12 +580,12 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False): trait.locus = trait.locus_chr = trait.locus_mb = trait.additive = "" else: trait.locus = trait.lrs = trait.additive = "" - - if (dataset.type == 'Publish' or dataset.type == "ProbeSet") and trait.locus_chr != "" and trait.locus_mb != "": - trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % (trait.locus_chr, float(trait.locus_mb)) - if trait.lrs != "": + if (dataset.type == 'Publish' or dataset.type == "ProbeSet") and str(trait.locus_chr or "") != "" and str(trait.locus_mb or "") != "": + trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % ( + trait.locus_chr, float(trait.locus_mb)) + if str(trait.lrs or "") != "": trait.LRS_score_repr = LRS_score_repr = '%3.1f' % trait.lrs else: - raise KeyError, `trait.name`+' information is not found in the database.' - - return trait
\ No newline at end of file + raise KeyError(repr(trait.name) + + ' information is not found in the database.') + return trait diff --git a/wqflask/base/webqtlCaseData.py b/wqflask/base/webqtlCaseData.py index d8487f01..25b6cb8a 100644 --- a/wqflask/base/webqtlCaseData.py +++ b/wqflask/base/webqtlCaseData.py @@ -19,8 +19,7 @@ # 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 utility.logger import getLogger logger = getLogger(__name__) @@ -29,58 +28,57 @@ import utility.tools utility.tools.show_settings() -class webqtlCaseData(object): + +class webqtlCaseData: """one case data in one trait""" def __init__(self, name, value=None, variance=None, num_cases=None, name2=None): self.name = name - self.name2 = name2 # Other name (for traits like BXD65a) + # Other name (for traits like BXD65a) + self.name2 = name2 self.value = value # Trait Value self.variance = variance # Trait Variance self.num_cases = num_cases # Number of individuals/cases self.extra_attributes = None - self.this_id = None # Set a sane default (can't be just "id" cause that's a reserved word) + # Set a sane default (can't be just "id" cause that's a reserved word) + self.this_id = None self.outlier = None # Not set to True/False until later def __repr__(self): - str = "<webqtlCaseData> " - if self.value != None: - str += "value=%2.3f" % self.value - if self.variance != None: - str += " variance=%2.3f" % self.variance + case_data_string = "<webqtlCaseData> " + if self.value is not None: + case_data_string += "value=%2.3f" % self.value + if self.variance is not None: + case_data_string += " variance=%2.3f" % self.variance if self.num_cases: - str += " ndata=%s" % self.num_cases + case_data_string += " ndata=%s" % self.num_cases if self.name: - str += " name=%s" % self.name + case_data_string += " name=%s" % self.name if self.name2: - str += " name2=%s" % self.name2 - return str + case_data_string += " name2=%s" % self.name2 + return case_data_string @property def class_outlier(self): """Template helper""" if self.outlier: return "outlier" - else: - return "" + return "" @property def display_value(self): - if self.value != None: + if self.value is not None: return "%2.3f" % self.value - else: - return "x" + return "x" @property def display_variance(self): - if self.variance != None: + if self.variance is not None: return "%2.3f" % self.variance - else: - return "x" + return "x" @property def display_num_cases(self): - if self.num_cases != None: + if self.num_cases is not None: return "%s" % self.num_cases - else: - return "x" + return "x" diff --git a/wqflask/base/webqtlConfig.py b/wqflask/base/webqtlConfig.py index 55407123..39947158 100644 --- a/wqflask/base/webqtlConfig.py +++ b/wqflask/base/webqtlConfig.py @@ -1,4 +1,4 @@ -#########################################' +# ' # Environment Variables - public # # Note: much of this needs to handled by the settings/environment @@ -10,31 +10,35 @@ from utility.tools import valid_path, mk_dir, assert_dir, assert_writable_dir, flat_files, TEMPDIR -#Debug Level -#1 for debug, mod python will reload import each time +# Debug Level +# 1 for debug, mod python will reload import each time DEBUG = 1 -#USER privilege -USERDICT = {'guest':1,'user':2, 'admin':3, 'root':4} +# USER privilege +USERDICT = {'guest': 1, 'user': 2, 'admin': 3, 'root': 4} -#minimum number of informative strains +# Set privileges +SUPER_PRIVILEGES = {'data': 'edit', 'metadata': 'edit', 'admin': 'edit-admins'} +DEFAULT_PRIVILEGES = {'data': 'view', 'metadata': 'view', 'admin': 'not-admin'} + +# minimum number of informative strains KMININFORMATIVE = 5 -#Daily download limit from one IP +# Daily download limit from one IP DAILYMAXIMUM = 1000 -#maximum LRS value +# maximum LRS value MAXLRS = 460.0 -#MINIMUM Database public value +# MINIMUM Database public value PUBLICTHRESH = 0 -#EXTERNAL LINK ADDRESSES +# EXTERNAL LINK ADDRESSES PUBMEDLINK_URL = "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=%s&dopt=Abstract" UCSC_BLAT = 'http://genome.ucsc.edu/cgi-bin/hgBlat?org=%s&db=%s&type=0&sort=0&output=0&userSeq=%s' UTHSC_BLAT = 'http://ucscbrowser.genenetwork.org/cgi-bin/hgBlat?org=%s&db=%s&type=0&sort=0&output=0&userSeq=%s' UTHSC_BLAT2 = 'http://ucscbrowserbeta.genenetwork.org/cgi-bin/hgBlat?org=%s&db=%s&type=0&sort=0&output=0&userSeq=%s' -GENOMEBROWSER_URL="https://genome.ucsc.edu/cgi-bin/hgTracks?db=%s&position=%s" +GENOMEBROWSER_URL = "https://genome.ucsc.edu/cgi-bin/hgTracks?db=%s&position=%s" NCBI_LOCUSID = "http://www.ncbi.nlm.nih.gov/gene?cmd=Retrieve&dopt=Graphics&list_uids=%s" GENBANK_ID = "http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Nucleotide&cmd=search&doptcmdl=DocSum&term=%s" OMIM_ID = "http://www.ncbi.nlm.nih.gov/omim/%s" @@ -52,26 +56,28 @@ GEMMA_URL = "http://www.chibi.ubc.ca/Gemma/gene/showGene.html?ncbiid=%s" ABA_URL = "http://mouse.brain-map.org/search/show?search_type=gene&search_term=%s" EBIGWAS_URL = "https://www.ebi.ac.uk/gwas/search?query=%s" WIKI_PI_URL = "http://severus.dbmi.pitt.edu/wiki-pi/index.php/search?q=%s" -ENSEMBLETRANSCRIPT_URL="http://useast.ensembl.org/Mus_musculus/Transcript/Idhistory?t=%s" +ENSEMBLETRANSCRIPT_URL = "http://useast.ensembl.org/Mus_musculus/Transcript/Idhistory?t=%s" DBSNP = 'http://ensembl.org/Mus_musculus/Variation/Population?v=%s' PROTEIN_ATLAS_URL = "http://www.proteinatlas.org/search/%s" OPEN_TARGETS_URL = "https://genetics.opentargets.org/gene/%s" UNIPROT_URL = "https://www.uniprot.org/uniprot/%s" RGD_URL = "https://rgd.mcw.edu/rgdweb/elasticResults.html?term=%s&category=Gene&species=%s" PHENOGEN_URL = "https://phenogen.org/gene.jsp?speciesCB=Rn&auto=Y&geneTxt=%s&genomeVer=rn6§ion=geneEQTL" +RRID_MOUSE_URL = "https://www.jax.org/strain/%s" +RRID_RAT_URL = "https://rgd.mcw.edu/rgdweb/report/strain/main.html?id=%s" # Temporary storage (note that this TMPDIR can be set as an # environment variable - use utility.tools.TEMPDIR when you # want to reach this base dir assert_writable_dir(TEMPDIR) -TMPDIR = mk_dir(TEMPDIR+'/gn2/') +TMPDIR = mk_dir(TEMPDIR + '/gn2/') assert_writable_dir(TMPDIR) -CACHEDIR = mk_dir(TMPDIR+'/cache/') +CACHEDIR = mk_dir(TMPDIR + '/cache/') # We can no longer write into the git tree: -GENERATED_IMAGE_DIR = mk_dir(TMPDIR+'generated/') -GENERATED_TEXT_DIR = mk_dir(TMPDIR+'generated_text/') +GENERATED_IMAGE_DIR = mk_dir(TMPDIR + 'generated/') +GENERATED_TEXT_DIR = mk_dir(TMPDIR + 'generated_text/') # Make sure we have permissions to access these assert_writable_dir(CACHEDIR) @@ -79,12 +85,12 @@ assert_writable_dir(GENERATED_IMAGE_DIR) assert_writable_dir(GENERATED_TEXT_DIR) # Flat file directories -GENODIR = flat_files('genotype')+'/' +GENODIR = flat_files('genotype') + '/' assert_dir(GENODIR) -assert_dir(GENODIR+'bimbam') # for gemma +# assert_dir(GENODIR+'bimbam') # for gemma # JSON genotypes are OBSOLETE -JSON_GENODIR = flat_files('genotype/json')+'/' +JSON_GENODIR = flat_files('genotype/json') + '/' if not valid_path(JSON_GENODIR): # fall back on old location (move the dir, FIXME) JSON_GENODIR = flat_files('json') @@ -92,4 +98,4 @@ if not valid_path(JSON_GENODIR): # Are we using the following...? PORTADDR = "http://50.16.251.170" INFOPAGEHREF = '/dbdoc/%s.html' -CGIDIR = '/webqtl/' #XZ: The variable name 'CGIDIR' should be changed to 'PYTHONDIR' +CGIDIR = '/webqtl/' # XZ: The variable name 'CGIDIR' should be changed to 'PYTHONDIR' |