diff options
Diffstat (limited to 'wqflask/base')
| -rw-r--r-- | wqflask/base/data_set.py | 44 | ||||
| -rw-r--r-- | wqflask/base/trait.py | 245 | ||||
| -rw-r--r-- | wqflask/base/webqtlConfig.py | 5 |
3 files changed, 157 insertions, 137 deletions
diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index cab708ef..2272b6ee 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -65,6 +65,9 @@ logger = getLogger(__name__ ) DS_NAME_MAP = {} 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) @@ -131,7 +134,7 @@ Publish or ProbeSet. E.g. ProbeSetFreeze.Name = "{0}" """.format(name) - results = g.db.execute(geno_query).fetchall() + results = g.db.execute(mrna_expr_query).fetchall() if len(results): self.datasets[name] = "ProbeSet" Redis.set("dataset_structure", json.dumps(self.datasets)) @@ -165,12 +168,11 @@ Publish or ProbeSet. E.g. geno_query = """ SELECT - GenoFreezeId + GenoFreeze.Id FROM GenoFreeze WHERE GenoFreeze.Name = "{0}" - {1} """.format(name) results = g.db.execute(geno_query).fetchall() @@ -487,25 +489,18 @@ class DatasetGroup(object): 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 @@ -513,12 +508,10 @@ 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]) @@ -638,29 +631,25 @@ 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') + 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)) @@ -669,9 +658,8 @@ 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)) diff --git a/wqflask/base/trait.py b/wqflask/base/trait.py index e454c593..8e11c11d 100644 --- a/wqflask/base/trait.py +++ b/wqflask/base/trait.py @@ -1,12 +1,12 @@ from __future__ import absolute_import, division, print_function +import os import string import resource import codecs import requests - -import redis -Redis = redis.StrictRedis() +import random +import urllib from base import webqtlConfig from base.webqtlCaseData import webqtlCaseData @@ -14,7 +14,10 @@ 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_VERSION +from utility.redis_tools import get_redis_conn, get_resource_id, get_resource_info +Redis = get_redis_conn() from wqflask import app @@ -22,11 +25,36 @@ import simplejson as json from MySQLdb import escape_string as escape from pprint import pformat as pf -from flask import Flask, g, request, url_for +from flask import Flask, g, request, url_for, redirect, make_response, render_template from utility.logger import getLogger logger = getLogger(__name__ ) +def create_trait(**kw): + assert bool(kw.get('dataset')) != bool(kw.get('dataset_name')), "Needs dataset ob. or name"; + + permitted = True + if kw.get('name'): + if kw.get('dataset_name'): + if kw.get('dataset_name') != "Temp": + dataset = create_dataset(kw.get('dataset_name')) + else: + dataset = kw.get('dataset') + + if kw.get('dataset_name') != "Temp": + if dataset.type == 'Publish': + permissions = check_resource_availability(dataset, kw.get('name')) + else: + permissions = check_resource_availability(dataset) + + if "view" in permissions['data']: + 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(object): """ Trait class defines a trait in webqtl, can be either Microarray, @@ -51,6 +79,7 @@ class GeneralTrait(object): self.haveinfo = kw.get('haveinfo', False) self.sequence = kw.get('sequence') # Blat sequence, available for ProbeSet self.data = kw.get('data', {}) + self.view = True # Sets defaults self.locus = None @@ -76,8 +105,6 @@ class GeneralTrait(object): # 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: self = retrieve_sample_data(self, self.dataset) @@ -117,6 +144,7 @@ class GeneralTrait(object): formatted = self.post_publication_description else: formatted = "Not available" + return formatted @property @@ -213,26 +241,28 @@ def get_sample_data(): 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() }]) + 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) + 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() }]) + else: + return None def jsonable(trait): """Return a dict suitable for using as json @@ -347,90 +377,94 @@ def jsonable_table_row(trait, dataset_name, index): 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 = "http://localhost:8080/run-action?resource={}&user={}&branch=data&action=view".format(resource_id, g.user_session.user_id) + else: + the_url = "http://localhost:8080/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, + 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 + WHERE + PublishXRef.Id = %s AND + Phenotype.Id = PublishXRef.PhenotypeId AND + Publication.Id = PublishXRef.PublicationId AND + PublishXRef.InbredSetId = PublishFreeze.InbredSetId 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 = 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() 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] + 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') + # holder = unicode(holder.strip(codecs.BOM_UTF8), 'utf-8', "ignore") + setattr(trait, field, holder) if dataset.type == 'Publish': @@ -449,13 +483,6 @@ def retrieve_trait_info(trait, dataset, get_qtl_info=False): if trait.confidential: 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 else: trait.abbreviation = trait.post_publication_abbreviation if description: diff --git a/wqflask/base/webqtlConfig.py b/wqflask/base/webqtlConfig.py index 018d5d54..3d86bc22 100644 --- a/wqflask/base/webqtlConfig.py +++ b/wqflask/base/webqtlConfig.py @@ -17,6 +17,10 @@ DEBUG = 1 #USER privilege USERDICT = {'guest':1,'user':2, 'admin':3, 'root':4} +#Set privileges +SUPER_PRIVILEGES = {'data': ['no-access', 'view', 'edit'], 'metadata': ['no-access', 'view', 'edit'], 'admin': ['not-admin', 'edit-access', 'edit-admins']} +DEFAULT_PRIVILEGES = {'data': ['no-access', 'view'], 'metadata': ['no-access', 'view'], 'admin': ['not-admin']} + #minimum number of informative strains KMININFORMATIVE = 5 @@ -58,6 +62,7 @@ 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" # Temporary storage (note that this TMPDIR can be set as an # environment variable - use utility.tools.TEMPDIR when you |