def add_one_sample(self, new_sample_dict):
if 'SAMPLE' not in new_sample_dict:
self.logger.info("The new sample entry does not contain a SAMPLE key.")
sys.exit()
else:
new_sample_dict.update({"PROJECT": self.project_config['project_name']})
sampleinfo_mongo.add_new_sample(new_sample_dict)
sample = new_sample_dict["SAMPLE"]
if not variants_mongo.is_sample_loaded(sample, "orig"):
self.__load_variants(sample)
def sample_variants_csv(self, sample, type):
if not sampleinfo_mongo.is_sample(sample) or not variants_mongo.is_sample_loaded(sample, type):
self.__log_sample_doesnt_exist()
return
out_path = "%s/%s.csv" % ( self.output_files_dir, sample)
print out_path
csv_writer = csv.writer(open(out_path, "w"), delimiter=',', quotechar='"', quoting=csv.QUOTE_NONNUMERIC)
header = ['CHROM', 'POS', 'REF', 'ALT', 'GT', 'RSID', 'Gene',
'ExonicFunc', 'AAChange', 'FREQ', 'QC_Final', 'QC_Cov', 'QC_AF', 'In_Hotspot']
csv_writer.writerow(header)
client, db = mongo.get_connection()
total_loaded_samples = variants_mongo.count_samples()
for var in variants_mongo.get_sample_vars(sample, type, db):
new_variant = {}
chrom, pos, ref, alt = var['CHROM'], var['POS'], var['REF'], var['ALT']
al1, al2 = genotypetools.get_genotype_alleles(ref, alt, var['GT_calc'])
new_variant.update({'CHROM': chrom, 'POS': pos, 'REF': ref, 'ALT': ",".join(alt),
'GT': "/".join([al1, al2])})
hotspot = hotspot_mongo.get_variant(chrom, pos, ref, alt, db)
annot = hotspot['ANNOTATION'][0]
new_variant.update({'RSID': annot['snp137NonFlagged'],
'Gene': annot['Gene_refGene'], 'ExonicFunc': annot['ExonicFunc_refGene'],
'AAChange': annot['AAChange_refGene']})
if 'p.' in new_variant['AAChange']:
new_variant['AAChange'] = new_variant['AAChange'].split('p.')[1].split(",")[0]
zygosity = hotspot['orig_stats']['zygosity']
freq = sum([zygosity['het_count'], zygosity['het_alt_count'], zygosity['hom_count']]) / float(total_loaded_samples)
final_qc, qc_cov, qc_af = var['FINAL_QC'], var['COV_QC'], var['AF_QC']
if hotspot['orig_stats']['qc']['final_qc_count'] > 0:
in_hotspot = "TRUE"
else:
in_hotspot = "FALSE"
new_variant.update({"FREQ": freq, "QC_Final": final_qc, "QC_Cov": qc_cov, "QC_AF": qc_af,
"In_Hotspot": in_hotspot})
out_row = [str(new_variant[field]) for field in header]
csv_writer.writerow(out_row)
#print "\t".join(out_row)
return out_path
def __get_unsaved_hotspot_vcf_files(self):
hotspot_dir = self.project_config['hotspot_dir']
output_dir = hotspot_dir + "/hotspot_output"
vcf_files = glob(output_dir+"/*.vcf")
final_vcf_files = []
client, db = mongo.get_connection()
for vcf_file in vcf_files:
sample = os.path.basename(vcf_file).split(".")[0]
if sampleinfo_mongo.is_sample(sample, db) and not \
variants_mongo.is_sample_loaded(sample, self.variant_type, db):
self.__log_adding_hotspot_sample_to_queue(sample, vcf_file)
final_vcf_files.append((sample, vcf_file))
else:
self.__log_hotspot_sample_already_loaded(sample)
client.close()
return final_vcf_files
def load_all(self):
if self.variant_type == 'orig':
client, db = mongo.get_connection()
vcf_files = sampleinfo_mongo.get_vcf_files()
# CHECK IF THE VCFS ARE ALL VALID BEFORE STARTING
for sample in vcf_files:
vcf_file = vcf_files[sample]
if not os.path.isfile(vcf_file):
self.__log_invalid_vcf_file(vcf_file)
sys.exit(1)
pending_vcf_files = []
for sample in vcf_files:
print sample
vcf_file = vcf_files[sample]
is_loaded = variants_mongo.is_sample_loaded(sample, self.variant_type, db)
if is_loaded:
self.__log_sample_already_loaded(sample)
continue
else:
self.__log_adding_sample_to_queue(sample, vcf_file)
pending_vcf_files.append((sample, vcf_file))
client.close()
elif self.variant_type == 'hotspot':
pending_vcf_files = self.__get_unsaved_hotspot_vcf_files()
num_processors = 10
self.__parallel_process_vcf_files(pending_vcf_files, num_processors)
self.__log_successfully_loaded()
