def rm2result(cls, session, rm, snp_info, max_rank=1000, commit=False, min_rank=1, results_directory=None):
"""
2009-11-2
split out of run()
"""
# 2009-5-1 check whether it's already in db.
db_entries = Stock_250kDB.Results.query.filter_by(results_id=rm.id)
if db_entries.count()==max_rank-min_rank+1:
sys.stderr.write("%s already in db. Ignore.\n"%rm.id)
return
param_data = PassingData(min_MAC=0)
genome_wide_result = GeneListRankTest.getResultMethodContent(rm, results_directory=results_directory, min_MAF=0., \
pdata=param_data)
if genome_wide_result:
for rank in range(min_rank, max_rank+1):
data_obj = genome_wide_result.get_data_obj_at_given_rank(rank)
if data_obj is not None:
snps_id = snp_info.getSnpsIDGivenChrPos(data_obj.chromosome, data_obj.position)
if data_obj.extra_col_ls:
result_obj = cPickle.dumps(data_obj.extra_col_ls)
else:
result_obj = None
Stock_250kDB.Results(snps_id=snps_id, results_id=rm.id, score=data_obj.value, rank=rank, beta=getattr(data_obj, 'beta1', None),\
maf=data_obj.maf, mac=data_obj.mac, genotype_var_perc=data_obj.genotype_var_perc,\
object=result_obj)
if commit:
session.flush()
#session.commit()
else:
session.rollback()
def run(self): if self.debug: import pdb pdb.set_trace() db = Stock_250kDB.Stock_250kDB(drivername=self.drivername, username=self.db_user, password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema) db.setup(create_tables=False) session = db.session #session.begin() snps_context_wrapper = GeneListRankTest.dealWithSnpsContextWrapper(self.snps_context_picklef, self.min_distance, self.get_closest) gene_annotation = DrawSNPRegion.dealWithGeneAnnotation(self.gene_annotation_picklef) snp_info = DrawSNPRegion.getSNPInfo(db) snp_annotation_short_name2id = self.getSNPAnnotationShortName2id() self._constructSNPAnnotation(session, snp_info, snps_context_wrapper, gene_annotation, snp_annotation_short_name2id) if self.commit: session.flush() session.commit()
def run(self): if self.debug: import pdb pdb.set_trace() db = Stock_250kDB.Stock_250kDB(drivername=self.drivername, username=self.db_user, password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema) db.setup(create_tables=False) session = db.session #session.begin() snps_context_wrapper = GeneListRankTest.dealWithSnpsContextWrapper(self.snps_context_picklef, self.min_distance, \ self.get_closest) gene_annotation = DrawSNPRegion.dealWithGeneAnnotation(self.gene_annotation_picklef, tax_id=self.tax_id, \ cls_with_db_args=self) snp_info = DrawSNPRegion.getSNPInfo(db) snp_annotation_short_name2id = self.getSNPAnnotationShortName2id() self._constructSNPAnnotation(session, snp_info, snps_context_wrapper, gene_annotation, snp_annotation_short_name2id) if self.commit: session.flush() session.commit()
def getTopNumberSNPs(self, call_method_id, phenotype_method_id, analysis_method_id=1, results_directory=None, min_MAC=7,\ no_of_top_snps=10000): """ 2009-3-18 if data_obj from genome_wide_result is None, ignore it 2009-2-16 get a certain number of top SNPs from a result (according to call_method_id, phenotype_method_id, analysis_method_id) """ rm = Stock_250kDB.ResultsMethod.query.filter_by(call_method_id=call_method_id).\ filter_by(phenotype_method_id=phenotype_method_id).filter_by(analysis_method_id=analysis_method_id).first() result_id = rm.id param_data = PassingData(min_MAC=min_MAC) genome_wide_result = GeneListRankTest.getResultMethodContent(rm, results_directory=results_directory, min_MAF=0., \ pdata=param_data) chr_pos_ls = [] for i in range(no_of_top_snps): data_obj = genome_wide_result.get_data_obj_at_given_rank(i+1) if data_obj is not None: chr_pos_ls.append((data_obj.chromosome, data_obj.position)) chr_pos_ls.sort() return chr_pos_ls
def run(self):
"""
"""
if self.debug:
import pdb
pdb.set_trace()
db = Stock_250kDB.Stock_250kDB(drivername=self.drivername, username=self.db_user,
password=self.db_passwd, hostname=self.hostname, database=self.dbname, schema=self.schema)
db.setup(create_tables=False)
session = db.session
#session.begin()
query = Stock_250kDB.ResultsMethod.query.filter_by(results_method_type_id=1)
if self.call_method_id:
query = query.filter_by(call_method_id=self.call_method_id)
for rm in query.all():
rows = Stock_250kDB.MAFVsScorePlot.query.filter_by(results_method_id=rm.id)
if rows.count()>0:
sys.stderr.write("Ignore. MAF vs score plot done for %s.\n"%rm.id)
continue
gwr = GeneListRankTest.getResultMethodContent(rm, results_directory=self.results_directory, min_MAF=0)
if gwr is None:
continue
if self.output_dir:
output_fname_prefix = os.path.join(self.output_dir, '%s_maf_vs_score'%rm.id)
else:
output_fname_prefix = None
png_data, svg_data = self.plot_maf_vs_score(rm, gwr, output_fname_prefix, commit=self.commit)
if png_data or svg_data:
maf_vs_score_plot = Stock_250kDB.MAFVsScorePlot(results_method_id=rm.id)
if png_data:
maf_vs_score_plot.png_data = png_data.getvalue()
if svg_data:
maf_vs_score_plot.svg_data = svg_data.getvalue()
session.save(maf_vs_score_plot)
del png_data, svg_data
if self.commit:
session.flush()
"""
def getTopNumberSNPs(self, call_method_id, phenotype_method_id, analysis_method_id=1, results_directory=None, min_MAC=7,\ no_of_top_snps=10000): """ 2009-3-18 if data_obj from genome_wide_result is None, ignore it 2009-2-16 get a certain number of top SNPs from a result (according to call_method_id, phenotype_method_id, analysis_method_id) """ rm = Stock_250kDB.ResultsMethod.query.filter_by(call_method_id=call_method_id).\ filter_by(phenotype_method_id=phenotype_method_id).filter_by(analysis_method_id=analysis_method_id).first() result_id = rm.id param_data = PassingData(min_MAC=min_MAC) genome_wide_result = GeneListRankTest.getResultMethodContent(rm, results_directory=results_directory, min_MAF=0., \ pdata=param_data) chr_pos_ls = [] for i in range(no_of_top_snps): data_obj = genome_wide_result.get_data_obj_at_given_rank(i+1) if data_obj is not None: chr_pos_ls.append((data_obj.chromosome, data_obj.position)) chr_pos_ls.sort() return chr_pos_ls
def on_button_draw_gene_list_bars_clicked(self, widget):
"""
2008-12-16
draw vertical spans to denote the locations of genes from a candidate list
"""
if self.db is None:
self.db_connect()
if not self.chr_id2size:
sys.stderr.write("No genome-wide pvalue plot has been drawn yet. Do it first!\n")
return
entry_gene_list_id = self.xml.get_widget("entry_gene_list_id")
list_type_id = entry_gene_list_id.get_text()
comboboxentry_bar_color = self.xml.get_widget("comboboxentry_bar_color")
bar_color = comboboxentry_bar_color.get_active_text()
if not bar_color: #default is black
bar_color = 'k'
if list_type_id:
list_type_id = int(list_type_id)
self.candidate_gene_set = GeneListRankTest.dealWithCandidateGeneList(list_type_id, return_set=True)
for gene_id in self.candidate_gene_set:
gene_model = self.gene_annotation.gene_id2model[gene_id]
if gene_id in self.gene_id2vspan_obj_id:
artist_obj_id = self.gene_id2vspan_obj_id[gene_id]
artist = self.artist_obj_id2artist_gene_id_ls[artist_obj_id][0]
if artist.get_edgecolor()!=bar_color:
artist.set_edgecolor(bar_color)
if artist.get_facecolor()!=bar_color:
artist.set_facecolor(bar_color)
#artist.remove()
else:
this_chr_starting_pos_on_plot = self.chr_id2cumu_size[gene_model.chromosome]-\
self.chr_id2size[gene_model.chromosome]-self.chr_gap
xmin = this_chr_starting_pos_on_plot + gene_model.start
xmax = this_chr_starting_pos_on_plot + gene_model.stop
artist = self.ax.axvspan(xmin, xmax, edgecolor=bar_color, facecolor=bar_color, alpha=0.3, picker=6)
artist_obj_id = id(artist)
self.artist_obj_id2artist_gene_id_ls[artist_obj_id] = [artist, gene_id]
self.gene_id2vspan_obj_id[gene_id] = artist_obj_id
self.canvas_matplotlib.draw()
def get_phenotype_ls(self, rm, no_of_top_snps, chr_pos2ancestral_allele, pheno_data, geno_data, min_MAF, results_directory=None):
"""
2008-09-26
consider the complement of the ancestral allele as ancestral as well.
2008-09-19
differentiate phenotype between ancestral/derived alleles
"""
sys.stderr.write("Getting phenotype_ls for ancestral/derived alleles ... ")
genome_wide_result = GeneListRankTest.getResultMethodContent(rm, results_directory, min_MAF)
genome_wide_result.data_obj_ls.sort() #in value descending order. each SNP object has a defined method for comparison based on its value
genome_wide_result.data_obj_ls.reverse()
ancestral_allele_phenotype_ls = []
derived_allele_phenotype_ls = []
no_of_genotype_accessions = len(geno_data.row_id_ls)
pheno_data_col_index = pheno_data.phenotype_method_id2col_index[rm.phenotype_method_id]
no_of_polarized_snps = 0
for i in range(no_of_top_snps):
data_obj = genome_wide_result.data_obj_ls[i]
chr_pos = '%s_%s'%(data_obj.chromosome, data_obj.position)
geno_data_col_index = geno_data.col_id2col_index.get(chr_pos)
if chr_pos in chr_pos2ancestral_allele and geno_data_col_index is not None:
no_of_polarized_snps += 1
for i in range(no_of_genotype_accessions):
allele = geno_data.data_matrix[i][geno_data_col_index]
pheno_row_index = pheno_data.row_id2row_index.get(geno_data.row_id_ls[i]) #find corresponding accession row index in phenotype matrix
if pheno_row_index is not None and allele>=1 and allele<=4: #no heterozygote or NA or deletion
phenotype = pheno_data.data_matrix[pheno_row_index][pheno_data_col_index]
if phenotype!='NA':
phenotype = float(phenotype)
if allele==chr_pos2ancestral_allele[chr_pos] or allele==number2complement[chr_pos2ancestral_allele[chr_pos]]:
ancestral_allele_phenotype_ls.append(phenotype)
else:
derived_allele_phenotype_ls.append(phenotype)
phenotype_ls_data = PassingData()
phenotype_ls_data.ancestral_allele_phenotype_ls = ancestral_allele_phenotype_ls
phenotype_ls_data.derived_allele_phenotype_ls = derived_allele_phenotype_ls
sys.stderr.write("no_of_polarized_snps/no_of_top_snps=%s/%s=%.3f. Done.\n"%(no_of_polarized_snps, no_of_top_snps, no_of_polarized_snps/float(no_of_top_snps)))
return phenotype_ls_data
def __init__(self, **keywords): """ 2008-08-20 """ GeneListRankTest.__init__(self, **keywords)
def __init__(self, **keywords): GeneListRankTest.__init__(self, **keywords) self.list_type_id_ls = getListOutOfStr(self.list_type_id_ls, data_type=int) self.analysis_method_id_ls = getListOutOfStr(self.analysis_method_id_ls, data_type=int) self.phenotype_method_id_ls = getListOutOfStr(self.phenotype_method_id_ls, data_type=int)
def __init__(self, **keywords): GeneListRankTest.__init__(self, **keywords) self.list_type_id_ls = getListOutOfStr(self.list_type_id_ls, data_type=int) self.analysis_method_id_ls = getListOutOfStr(self.analysis_method_id_ls, data_type=int) self.phenotype_method_id_ls = getListOutOfStr(self.phenotype_method_id_ls, data_type=int)
def __init__(self, **keywords): """ 2008-08-20 """ GeneListRankTest.__init__(self, **keywords)