#!/usr/bin/env python

import csv
import re
import argparse
import os
import sys


def __main__():

   parser = argparse.ArgumentParser(description='Transform FACETS output file to a format usable for input to PCGR')
   parser.add_argument('input_facets_tsv', help='FACETS tsv file with copy number segments ')
   parser.add_argument('output_pcgr_tsv', help='Tab-separated file of CNA segments formatted according to PCGR input requirements')
   parser.add_argument('sample_id', help='sample identifier')
   args = parser.parse_args()

   facets_to_pcgr(args.input_facets_tsv, args.output_pcgr_tsv, args.sample_id)

def facets_to_pcgr(input_tsv, output_tsv, sample_id):

   o = open(output_tsv,'w')
   o.write('Chromosome\tStart\tEnd\tSegment_Mean\tCellular_Fraction\tTotal_CN\tMinor_CN\tSampleID\n')
   with open(input_tsv, 'r') as tsvfile:
      reader = csv.DictReader(tsvfile, delimiter='\t')
      for rec in reader:
         chromosome = rec['#chromosome']
         if chromosome == '23':
            chromosome = 'X'
         if chromosome == '24':
            chromosome = 'Y'
         segment_start = str(rec['segment_start'])
         segment_end = str(rec['segment_end'])
         if not segment_start.isdigit() or not segment_end.isdigit():
            print('Unable to parse segment on chromosome - ' + str(chromosome) + ': segment_start (' + str(segment_start) + ') or segment_end (' + str(segment_end) + ') not proper integers')
         else:
            log2_ratio = rec['CN_logR_median_in_cluster']
            cf = rec['cluster_specific_cellular_fraction_MM_EM_optimized']
            total_cn = rec['cluster_specific_total_copy_number_MM_EM_optimized']
            minor_cn = rec['cluster_specific_minor_copy_number_MM_EM_optimized']

            o.write(str(chromosome) + '\t' + str(segment_start) + '\t' + str(segment_end) + '\t' + str(log2_ratio) + '\t' + str(cf) + '\t' + str(total_cn) + '\t' + str(minor_cn) + '\t' + str(sample_id) + '\n')

   o.close()

if __name__=="__main__": __main__()