#!/usr/bin/perl
# Description: run tRNAscan SE on input fasta
# Author: Peter Fischer Hallin
# Email: pfh@cbs.dtu.dk
# Version: Genome Atlas 3.0 ws2
# Date: 2009-09-02

# Reads a genomic sequence from STDIN and submits to tRNAscan . R-value cutoff, and model are
# defined as argv 0 and argv 1
# output is orfs

# include standard XML::Compile helper functions (used to initiate WSDL proxys)
require "xml-compile.pl";
require "fasta.inc.pl";


die "usage: trnascan.pl bac/arc/euk < genome.fsa\n" unless defined $ARGV[0];

use Data::Dumper;
my $trnascan = WSDLclient ( 'http://www.cbs.dtu.dk/ws/GenomeAtlas/GenomeAtlas_3_0_ws2.wsdl' );

my @fasta = read_fasta();

for ( my $i = 0 ; $i < scalar ( @fasta ) ; $i++ ) {
	my $id = $fasta[$i]->{id};
	my $comment = $fasta[$i]->{comment};
	my $seq = $fasta[$i]->{seq};
	run_trnascan  ($id,$comment,$seq) ;
}

sub run_trnascan {
	my ($jobid,$status,$expires);
	my ($id,$comment,$seq) = @_;
	my $response;
	$response = $trnascan->{trnascanRun}->(
 parameters => {
  parameters => {
   kingdom => $ARGV[0] ,
   sequence => { id => $id , comment => $comment , seq => $seq}}});
		die "error obtaining jobid\n" unless defined $response->{queueentry}->{queueentry};
		$jobid = $response->{queueentry}->{queueentry}->{jobid};

		wait_job($trnascan->{pollQueue},$jobid);

		my $response = $trnascan->{trnascanFetchResult}->( job => { jobid => $jobid   }) ;
	print  STDERR "# parsing 'anndata' object\n";
	foreach my $ann (@{$response->{parameters}->{anndata}->{ann}}) {
		printf STDERR "# annotations from %s version %s\n" , 
		  $response->{parameters}->{anndata}->{annsource}->{method} , 
		  $response->{parameters}->{anndata}->{annsource}->{version};
		foreach my $record (@{$ann->{annrecords}->{annrecord}}) {
			my ( $begin , $end ) = ( $record->{range}->{begin}->numify , $record->{range}->{end}->numify );
			my $score = sprintf("%0.4f",$record->{score}[0]->{value}->numify);
			my $new_id = "";
			my $feature = $record->{feature};
			if ( $begin < $end)  {
				my $len = ( $end - $begin) + 1;
				$gene = substr($seq,$begin-1,$len);
			} else {
				my $len = ( $begin - $end) + 1;
				$gene = substr($seq,$end-1, $len);	
				$gene = rev($gene);
			}
			print ">$id\_tRNA_$begin-$end /score=$score $comment\n";
			for (my $x=0 ; $x < length ( $gene ) ; $x += 60) {
				print substr($gene,$x,60),"\n";
			}
		}
	}
}

sub rev	{
	my $output = reverse $_[0];
	$output =~ tr/ACTGactg/TGACtgac/;
	return $output;
}