Stream VCF file from AWS s3 and do stuff (sort, gzip, index, subset for specific region)

Stream VCF from S3

#!/usr/bin/bash
#
# make_gz.sh
#

# Call this script with a list of s3 locations with VCF files to parse
# aws --profile NDAR s3 ls s3:/S3_URL/ | awk '{print $4}' | xargs -n1 -P4 sh make_gz.sh
# xargs -n1 -P4 accepts one argument and runs 4 parallel processes
#

#Create named pipe
mkfifo $1_pipe

#Set up stream for pipe
aws --profile NDAR s3 cp s3://S3_URL/$1 - | /usr/bin/vcftools/vcftools_0.1.11/perl/vcf-sort -c > $1_pipe &

#Use pipe output to create bgzip file
/usr/bin/htslib/htslib/bgzip -c $1_pipe > $1.gz

#Index bgzip format file
/usr/bin/htslib/bcftools/bcftools index $1.gz

#Remove named pipe
rm $1_pipe

#Query VCF for UBE3A gene location, output VCF, gzip, and index.
/usr/bin/htslib/bcftools/bcftools view --regions 15:25337244-25439042 $1.gz | /usr/bin/htslib/htslib/bgzip -c > $1.query.gz
/usr/bin/htslib/bcftools/bcftools index $1.query.gz

#Remove original vcf and index files
rm $1.gz
rm $1.gz.csi