def parse_input_file(TF, connection, input_file, gencode_file, fdr_filter=None, use_variance=None, sample_size=None, only_best_snp=None):
genes = {}
logging.info("Opening pheno phile")
with gzip.open(input_file) as file:
for i,line in enumerate(file):
if i==0:
continue
comps = line.strip().split()
if fdr_filter:
fdr = comps[TF.FDR]
if float(fdr) > fdr_filter:
snp = comps[TF.SNPName]
logging.log(9,"Snp %s doesn't pass fdr filter: %s", snp, fdr)
continue
gene = comps[TF.HUGO]
if "," in gene:
multiple_genes = gene.split(",")
for g in multiple_genes:
row = row_from_comps(g, comps, TF)
process_row(g, row, genes, only_best_snp)
else:
row = row_from_comps(gene, comps, TF)
process_row(gene, row, genes, only_best_snp)
logging.info("Opening gencode file")
class GenCodeCallback(object):
def __init__(self, genes):
self.genes = genes
self.selected = {}
def __call__(self, gencode):
if gencode.name in self.genes:
rows = self.genes[gencode.name]
self.selected[gencode.name] = [(row[0], gencode.ensemble_version, row[1], row[2], row[3], row[4]) for row in rows]
callback = GenCodeCallback(genes)
GencodeFile.parse_gencode_file(gencode_file, callback)
genes = callback.selected
if use_variance:
logging.info("Opening variance file")
vars = VarianceFile.load_variance(use_variance)
keys = genes.keys()
for key in keys:
rows = genes[key]
new_rows = []
for r in rows:
snp = r[0]
if not snp in vars:
continue
v = vars[snp]
std = math.sqrt(v/sample_size)
new_rows.append([r[0], r[1], r[2], str(float(r[3])*std), r[4], r[5]])
genes[key] = new_rows
Utilities.insert_entries(connection, genes)
def run(self):
if os.path.exists(self.args.output_folder):
logging.info("Output folder already exists, delete it if you want it done again")
return
os.makedirs(self.args.output_folder)
logging.info("Loading variance")
variances = VarianceFile.load_variance(self.args.variance_path)
logging.info("Processing GWAS")
index, results = gather_pb8k(self.args.pb8k_path, variances, self.args.sample_size)
logging.info("Writing slices")
for gene, rows in results.iteritems():
save_slice(self.args.output_folder, gene, rows)
def __init__(self, variance_file_path, sample_size, TF=TF1):
self.TF = TF
logging.info("Opening variance file")
self.vars = VarianceFile.load_variance(variance_file_path)
self.sample_size = sample_size
