parseScanMotifsOutput

#!/usr/bin/env python

# Parse the output of scanRBPmotifs and reports a string of all the RNBP hits for each gene_ID.
# The output facilitates the heterogeneous clustering.
#
# June 2018, UMR7216 Paris Diderot
# by Costas Bouyioukos cbouyio@gmail.com

import argparse
import sys


parser = argparse.ArgumentParser(description = "Parse scanRBPMotifs")
parser.add_argument("infh", type = argparse.FileType('r'), metavar ="input_file", help =  "Output table of scanRBPmotifs")
parser.add_argument("outfh", type = argparse.FileType('w'), metavar = "output_file", default = sys.stdout, help = "A table with <gene_ID> <text> fields")

optsArgs = parser.parse_args()

allLines = optsArgs.infh.readlines()

geneRBP = {}

for i in xrange(len(allLines)):
  fields = allLines[i].split("\t")
  gene = str(fields[9])
  rbp = str(fields[4])
  if gene in geneRBP:
    geneRBP[gene].append(rbp)
  else:
    geneRBP[gene] = [rbp]

for key, value in geneRBP.items():
  sortedV = sorted(value, key=str.lower)
  geneRBP[key] = sortedV

for key, value in geneRBP.items():
  rbps = "".join(value)
  string = key + "\t" + rbps
  print >> optsArgs.outfh, string