csv_from_final_reports.sh

#!/bin/sh
#$ -S /bin/bash
#$ -cwd
# Christian Theil HAve, 2015

# Modify this to be the final report directory of interest 
# Note that final reports must contain LRR and BAF
FINAL_REPORTS_DIR=target_final_reports

# This is where 
BAF_LRR_CSV_DIR=`pwd`/baf_lrr_csv



function nth() {
  target=$1
  shift
  idx=1
  for symbol in $@; do
    if [ $symbol == $target ]; then
      echo $idx
      return
    fi
    idx=$(($idx+1))
  done
}

rm -f samples_csv.list

# Create .csv files for each final report
for fr in $FINAL_REPORTS_DIR/*
do
	hdr_line=`tail -n+11 $fr | head -n1|tr '\t' ','`
	#echo $hdr_line
	IFS=,
	baf_idx=`nth "B Allele Freq" $hdr_line`
	lrr_idx=`nth "Log R Ratio" $hdr_line`
	sample_idx=`nth "Sample ID" $hdr_line`
	IFS=
	sample_id=$(tail -n+12 $fr | head -n1 |awk -F $'\t' "BEGIN {OFS = FS} {print \$${sample_idx}}")
	particid=`grep -w $sample_id BarcodesToParticidAllChips.txt | cut -d, -f2`
	if [ `grep -c -w "$particid" /eva/data/clean/target/chip_data/Target_and_GBparents_clean_V1.fam` = 1 ]; then
		echo "Processing sample $sample_id ($particid) from $fr"
		tail -n+12 $fr | awk -F $'\t' "BEGIN {OFS = FS} {print \$${baf_idx} \",\" \$${lrr_idx}}"  > $BAF_LRR_CSV_DIR/$particid.csv
		echo "$BAF_LRR_CSV_DIR/$particid.csv" >> samples_csv.list
	else
		echo "Skipping sample $sample_id ($particid) from $fr"
	fi
		

done