diff options
Diffstat (limited to 'wqflask/base/trait.py')
| -rwxr-xr-x | wqflask/base/trait.py | 96 |
1 files changed, 61 insertions, 35 deletions
diff --git a/wqflask/base/trait.py b/wqflask/base/trait.py index 241bf2ab..82e013ae 100755 --- a/wqflask/base/trait.py +++ b/wqflask/base/trait.py @@ -1,6 +1,8 @@ from __future__ import absolute_import, division, print_function import string +import resource + from htmlgen import HTMLgen2 as HT @@ -15,22 +17,38 @@ from pprint import pformat as pf from flask import Flask, g -class GeneralTrait: +def print_mem(stage=""): + mem = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss + print("{}: {}".format(stage, mem/1024)) + +class GeneralTrait(object): """ Trait class defines a trait in webqtl, can be either Microarray, Published phenotype, genotype, or user input trait """ - def __init__(self, **kw): - print("in GeneralTrait") - self.dataset = kw.get('dataset') # database name + def __init__(self, get_qtl_info=False, **kw): + # xor assertion + assert bool(kw.get('dataset')) != bool(kw.get('dataset_name')), "Needs dataset ob. xor name"; + if kw.get('dataset_name'): + self.dataset = create_dataset(kw.get('dataset_name')) + else: + self.dataset = kw.get('dataset') self.name = kw.get('name') # Trait ID, ProbeSet ID, Published ID, etc. 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 self.data = kw.get('data', {}) + + # Sets defaultst + self.locus = None + self.lrs = None + self.pvalue = None + self.mean = None + self.num_overlap = None + if kw.get('fullname'): name2 = value.split("::") @@ -39,13 +57,12 @@ class GeneralTrait: # self.cellid is set to None above elif len(name2) == 3: self.dataset, self.name, self.cellid = name2 - - self.dataset = create_dataset(self.dataset) # 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 - self.retrieve_info() + self.retrieve_info(get_qtl_info=get_qtl_info) self.retrieve_sample_data() + def get_name(self): @@ -78,7 +95,7 @@ class GeneralTrait: #desc = self.handle_pca(desc) stringy = desc return stringy - + def display_name(self): @@ -208,7 +225,7 @@ class GeneralTrait: # ''' % (self.cellid, self.name, self.dataset.name) # #else: - results = self.dataset.retrieve_sample_data(self) + results = self.dataset.retrieve_sample_data(self.name) # Todo: is this necessary? If not remove self.data.clear() @@ -229,7 +246,7 @@ class GeneralTrait: #def items(self): # return self.__dict__.items() - def retrieve_info(self, QTL=False): + def retrieve_info(self, get_qtl_info=False): assert self.dataset, "Dataset doesn't exist" if self.dataset.type == 'Publish': query = """ @@ -269,7 +286,7 @@ class GeneralTrait: escape(self.dataset.name), escape(self.name)) traitInfo = g.db.execute(query).fetchone() - print("traitInfo is: ", pf(traitInfo)) + #print("traitInfo is: ", pf(traitInfo)) #XZ, 05/08/2009: We also should use Geno.Id to find marker instead of just using Geno.Name # to avoid the problem of same marker name from different species. elif self.dataset.type == 'Geno': @@ -287,7 +304,7 @@ class GeneralTrait: escape(self.dataset.name), escape(self.name)) traitInfo = g.db.execute(query).fetchone() - print("traitInfo is: ", pf(traitInfo)) + #print("traitInfo is: ", pf(traitInfo)) else: #Temp type query = """SELECT %s FROM %s WHERE Name = %s """ % (string.join(self.dataset.display_fields,','), @@ -314,32 +331,32 @@ class GeneralTrait: #XZ, 05/26/2010: From time to time, this query get error message because some geneid values in database are not number. #XZ: So I have to test if geneid is number before execute the query. #XZ: The geneid values in database should be cleaned up. - try: - junk = float(self.geneid) - geneidIsNumber = 1 - except: - geneidIsNumber = 0 - - if geneidIsNumber: - query = """ - SELECT - HomologeneId - FROM - Homologene, Species, InbredSet - WHERE - Homologene.GeneId =%s AND - InbredSet.Name = '%s' AND - InbredSet.SpeciesId = Species.Id AND - Species.TaxonomyId = Homologene.TaxonomyId - """ % (escape(str(self.geneid)), escape(self.dataset.group.name)) - result = g.db.execute(query).fetchone() - else: - result = None + #try: + # float(self.geneid) + # geneidIsNumber = True + #except ValueError: + # geneidIsNumber = False + + #if geneidIsNumber: + query = """ + SELECT + HomologeneId + FROM + Homologene, Species, InbredSet + WHERE + Homologene.GeneId =%s AND + InbredSet.Name = '%s' AND + InbredSet.SpeciesId = Species.Id AND + Species.TaxonomyId = Homologene.TaxonomyId + """ % (escape(str(self.geneid)), escape(self.dataset.group.name)) + result = g.db.execute(query).fetchone() + #else: + # result = None if result: self.homologeneid = result[0] - if QTL: + if get_qtl_info: if self.dataset.type == 'ProbeSet' and not self.cellid: traitQTL = g.db.execute(""" SELECT @@ -355,8 +372,17 @@ class GeneralTrait: #traitQTL = self.cursor.fetchone() if traitQTL: self.locus, self.lrs, self.pvalue, self.mean = traitQTL + if self.locus: + result = g.db.execute(""" + select Geno.Chr, Geno.Mb from Geno, Species + where Species.Name = '%s' and + Geno.Name = '%s' and + Geno.SpeciesId = Species.Id + """, (species, self.locus)).fetchone() + self.locus_chr = result[0] + self.locus_mb = result[1] else: - self.locus = self.lrs = self.pvalue = self.mean = "" + self.locus = self.locus_chr = self.locus_mb = self.lrs = self.pvalue = self.mean = "" if self.dataset.type == 'Publish': traitQTL = g.db.execute(""" SELECT |