Extract SNPs from gen files

extractSnpsGenfiles.sh

#!/bin/bash
 
set -e


# This script extracts a list of SNPs from gzipped .gen files that are split into 23 chromosomes
# and outputs as a single gzipped .gen file containing all extracted SNPs that it can find
# To run use the command
#
# $ ./extractSnpsGenfiles.sh \
#		<snplistfile> \
#		/panfs/panasas01/shared/alspac/alspac_combined_1kg_20140424/chrCHR/alspac_1kg_p1v3_CHR_fixed.gz \
#		/panfs/panasas01/shared/alspac/alspac_combined_1kg_20140424/chrCHR/alspac_1kg_p1v3_CHR.sample \
#		<outputfile>
#
# where <snplistfile> is a text file with one SNP name per line and <outputfile> is the rootname of the output.
# e.g if <outputfile> is extractedsnps then it will make two new files - extractedsnps.gz and extractedsnps.sample
#
# WARNING!! This script will overwrite existing files called <outputfile>.gz and <outputfile>.sample

 
snplistfile=${1}
genfile=${2}
samplefile=${3}
outfile=${4}

touch ${outfile}.gz
rm ${outfile}.gz
touch ${outfile}.gz

for i in {1..23}
do
	filename1=$(sed -e "s/CHR/$i/g" <<< ${genfile})
	filename2=$(sed -e "s/CHR/$i/g" <<< ${samplefile})
	echo ""
	echo "$filename1"
	echo "$filename2"
	echo "${outfile}_${i}"
	echo ""
	gtool -S \
		--g ${filename1} \
		--s ${filename2} \
		--inclusion ${snplistfile} \
		--og ${outfile}_${i} \
		--os ${outfile}_${i}.sample 
	echo "$?"
	if [ -f "${outfile}_${i}.gz" ]; then
		cat "${outfile}_${i}.gz" >> ${outfile}.gz
		cp ${outfile}_${i}.sample ${outfile}.sample
	fi
done

# gtool -M --g ${genfilelist} --s ${samplefilelist} --og ${outfile} --os ${outfile}.sample
# gzip ${outfile}

rm ${outfile}_*