make-igv-genome-json.pl

make-igv-genome-json.pl

#!/usr/bin/env perl
# make-igv-genome-json.pl
# Peter Menzel
#
# This script creates an IGV genome json file (https://github.com/igvteam/igv/wiki/JSON-Genome-Format)
# The arguments are the paths to fasta, fai and gff files.
#
# Example usage:
# datasets download genome accession GCF_002101575.1
# unzip ncbi_dataset.zip
# samtools faidx ncbi_dataset/data/GCF_002101575.1/GCF_002101575.1_ASM210157v1_genomic.fna
# make-igv-genome-json.pl ncbi_dataset/data/GCF_002101575.1/GCF_002101575.1_ASM210157v1_genomic.fna.fai ncbi_dataset/data/GCF_002101575.1/GCF_002101575.1_ASM210157v1_genomic.fna ncbi_dataset/data/GCF_002101575.1/genomic.gff
#

use strict;
use warnings;

my $id;
my $fasta;
my $fai;
my %tracks;

while(my $arg = pop(@ARGV)) {
	die "$arg is not a readable file." unless -r $arg and ! -d $arg;
	if($arg =~ /.*?([^\/]+)\.(fasta|fna|fa)$/) {
		$fasta = $arg;
		$id = $1;
	}
	elsif($arg =~ /\.fai$/) {
		$fai = $arg;
	}
	elsif($arg =~ /\.(gff|gtf|gff3)$/) {
		$tracks{$arg} = $1;
	}
}
die "Missing FASTA file (.fa, .fna, or .fasta)" unless length $fasta;
die "Missing FASTA Index (.fai) file" unless length $fai;

print "{\n  \"id\": \"$id\",\n  \"name\": \"$id\",\n  \"fastaURL\": \"$fasta\",\n  \"indexURL\": \"$fai\"";
if(length keys %tracks) {
	print ",\n  \"tracks\": [\n";
	my @list_track_ids = keys %tracks;
	foreach my $i (0..$#list_track_ids) {
		my $type = $tracks{$list_track_ids[$i]};
		print "    {\n      \"name\": \"Genes ", $type, "\",\n      \"url\": \"$list_track_ids[$i]\",\n      \"format\": \"$type\"\n    }";
		print $i == $#list_track_ids ? "\n" : ",\n";
	}
	print "  ]\n}\n";
}