diff options
Diffstat (limited to 'wqflask/base/data_set.py')
-rw-r--r-- | wqflask/base/data_set.py | 147 |
1 files changed, 3 insertions, 144 deletions
diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index a4eaaa2e..9ca880d0 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -29,7 +29,6 @@ import json import gzip import cPickle as pickle import itertools -from operator import itemgetter from redis import Redis Redis = Redis() @@ -316,9 +315,6 @@ class DatasetGroup(object): return mapping_id, mapping_names - def get_specified_markers(self, markers = []): - self.markers = HumanMarkers(self.name, markers) - def get_markers(self): logger.debug("self.species is:", self.species) @@ -449,7 +445,6 @@ def datasets(group_name, this_group = None): group_name, webqtlConfig.PUBLICTHRESH, "'" + group_name + "'", webqtlConfig.PUBLICTHRESH)) - #for tissue_name, dataset in itertools.groupby(the_results, itemgetter(0)): for dataset_item in the_results: tissue_name = dataset_item[0] dataset = dataset_item[1] @@ -457,14 +452,10 @@ def datasets(group_name, this_group = None): if tissue_name in ['#PublishFreeze', '#GenoFreeze']: dataset_menu.append(dict(tissue=None, datasets=[(dataset, dataset_short)])) else: - dataset_sub_menu = [item[1:] for item in dataset] - tissue_already_exists = False - tissue_position = None for i, tissue_dict in enumerate(dataset_menu): if tissue_dict['tissue'] == tissue_name: tissue_already_exists = True - tissue_position = i break if tissue_already_exists: @@ -719,20 +710,6 @@ class PhenotypeDataSet(DataSet): # (Urgently?) Need to write this pass - def get_trait_list(self): - query = """ - select PublishXRef.Id - from PublishXRef, PublishFreeze - where PublishFreeze.InbredSetId=PublishXRef.InbredSetId - and PublishFreeze.Id = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - def get_trait_info(self, trait_list, species = ''): for this_trait in trait_list: @@ -746,7 +723,7 @@ class PhenotypeDataSet(DataSet): #of the post-publication description if this_trait.confidential: this_trait.description_display = "" - continue # for now + continue # for now, because no authorization features if not webqtlUtil.hasAccessToConfidentialPhenotypeTrait( privilege=self.privilege, @@ -770,9 +747,7 @@ class PhenotypeDataSet(DataSet): #LRS and its location this_trait.LRS_score_repr = "N/A" - this_trait.LRS_score_value = 0 this_trait.LRS_location_repr = "N/A" - this_trait.LRS_location_value = 1000000 if this_trait.lrs: query = """ @@ -789,17 +764,7 @@ class PhenotypeDataSet(DataSet): LRS_Chr = result[0] LRS_Mb = result[1] - #XZ: LRS_location_value is used for sorting - try: - LRS_location_value = int(LRS_Chr)*1000 + float(LRS_Mb) - except: - if LRS_Chr.upper() == 'X': - LRS_location_value = 20*1000 + float(LRS_Mb) - else: - LRS_location_value = ord(str(LRS_chr).upper()[0])*1000 + float(LRS_Mb) - this_trait.LRS_score_repr = LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_score_value = LRS_score_value = this_trait.lrs this_trait.LRS_location_repr = LRS_location_repr = 'Chr%s: %.6f' % (LRS_Chr, float(LRS_Mb)) def retrieve_sample_data(self, trait): @@ -861,40 +826,13 @@ class GenotypeDataSet(DataSet): def check_confidentiality(self): return geno_mrna_confidentiality(self) - def get_trait_list(self): - query = """ - select Geno.Name - from Geno, GenoXRef - where GenoXRef.GenoId = Geno.Id - and GenoFreezeId = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - def get_trait_info(self, trait_list, species=None): for this_trait in trait_list: if not this_trait.haveinfo: this_trait.retrieveInfo() - #XZ: trait_location_value is used for sorting - trait_location_repr = 'N/A' - trait_location_value = 1000000 - if this_trait.chr and this_trait.mb: - try: - trait_location_value = int(this_trait.chr)*1000 + this_trait.mb - except: - if this_trait.chr.upper() == 'X': - trait_location_value = 20*1000 + this_trait.mb - else: - trait_location_value = ord(str(this_trait.chr).upper()[0])*1000 + this_trait.mb - this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, float(this_trait.mb) ) - this_trait.location_value = trait_location_value def retrieve_sample_data(self, trait): query = """ @@ -989,20 +927,6 @@ class MrnaAssayDataSet(DataSet): def check_confidentiality(self): return geno_mrna_confidentiality(self) - def get_trait_list_1(self): - query = """ - select ProbeSet.Name - from ProbeSet, ProbeSetXRef - where ProbeSetXRef.ProbeSetId = ProbeSet.Id - and ProbeSetFreezeId = {} - """.format(escape(str(self.id))) - logger.sql(query) - results = g.db.execute(query).fetchall() - trait_data = {} - for trait in results: - trait_data[trait[0]] = self.retrieve_sample_data(trait[0]) - return trait_data - def get_trait_info(self, trait_list=None, species=''): # Note: setting trait_list to [] is probably not a great idea. @@ -1034,27 +958,8 @@ class MrnaAssayDataSet(DataSet): # Save it for the jinja2 template this_trait.description_display = description_display - #XZ: trait_location_value is used for sorting - trait_location_repr = 'N/A' - trait_location_value = 1000000 - if this_trait.chr and this_trait.mb: - #Checks if the chromosome number can be cast to an int (i.e. isn't "X" or "Y") - #This is so we can convert the location to a number used for sorting - trait_location_value = self.convert_location_to_value(this_trait.chr, this_trait.mb) - #try: - # trait_location_value = int(this_trait.chr)*1000 + this_trait.mb - #except ValueError: - # if this_trait.chr.upper() == 'X': - # trait_location_value = 20*1000 + this_trait.mb - # else: - # trait_location_value = (ord(str(this_trait.chr).upper()[0])*1000 + - # this_trait.mb) - - #ZS: Put this in function currently called "convert_location_to_value" - this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, - float(this_trait.mb)) - this_trait.location_value = trait_location_value + this_trait.location_repr = 'Chr%s: %.6f' % (this_trait.chr, float(this_trait.mb)) #Get mean expression value query = ( @@ -1076,9 +981,7 @@ class MrnaAssayDataSet(DataSet): #LRS and its location this_trait.LRS_score_repr = 'N/A' - this_trait.LRS_score_value = 0 this_trait.LRS_location_repr = 'N/A' - this_trait.LRS_location_value = 1000000 #Max LRS and its Locus location if this_trait.lrs and this_trait.locus: @@ -1093,40 +996,10 @@ class MrnaAssayDataSet(DataSet): if result: lrs_chr, lrs_mb = result - #XZ: LRS_location_value is used for sorting - lrs_location_value = self.convert_location_to_value(lrs_chr, lrs_mb) this_trait.LRS_score_repr = '%3.1f' % this_trait.lrs - this_trait.LRS_score_value = this_trait.lrs this_trait.LRS_location_repr = 'Chr%s: %.6f' % (lrs_chr, float(lrs_mb)) - - def convert_location_to_value(self, chromosome, mb): - try: - location_value = int(chromosome)*1000 + float(mb) - except ValueError: - if chromosome.upper() == 'X': - location_value = 20*1000 + float(mb) - else: - location_value = (ord(str(chromosome).upper()[0])*1000 + - float(mb)) - - return location_value - - def get_sequence(self): - query = """ - SELECT - ProbeSet.BlatSeq - FROM - ProbeSet, ProbeSetFreeze, ProbeSetXRef - WHERE - ProbeSet.Id=ProbeSetXRef.ProbeSetId and - ProbeSetFreeze.Id = ProbeSetXRef.ProbSetFreezeId and - ProbeSet.Name = %s - ProbeSetFreeze.Name = %s - """ % (escape(self.name), escape(self.dataset.name)) - logger.sql(query) - results = g.db.execute(query).fetchone() - return results[0] + return trait_list def retrieve_sample_data(self, trait): query = """ @@ -1150,7 +1023,6 @@ class MrnaAssayDataSet(DataSet): #logger.debug("RETRIEVED RESULTS HERE:", results) return results - def retrieve_genes(self, column_name): query = """ select ProbeSet.Name, ProbeSet.%s @@ -1204,19 +1076,6 @@ class TempDataSet(DataSet): desc = self.handle_pca(desc) return desc - def get_group(self): - query = """ - SELECT - InbredSet.Name, InbredSet.Id - FROM - InbredSet, Temp - WHERE - Temp.InbredSetId = InbredSet.Id AND - Temp.Name = "%s" - """ % self.name - logger.sql(query) - self.group, self.group_id = g.db.execute(query).fetchone() - def retrieve_sample_data(self, trait): query = """ SELECT |