#!/usr/local/bin/perl
#
# read gene association file from SGD GO and output two
# sequence files: nucleic acid and amino acid.
#
#

use strict;

my $external_table = "/share/ftp/yeast/data_dump/external_id/external_id.tab";
my $output = ">gp2protein.sgd";
my %records;
my %swiss;

open (EXT_IN, "<$external_table") || die "Cannot open $external_table\n";

while (<EXT_IN>) {
    chop;

    my ($extid, $type, $orfname, $sgdid) = split('\t');

    if ($type =~ /SwissProt/) {
	$swiss{$sgdid} = $extid;
    }
}

close (EXT_IN);

open (GP2P, "$output") || die "Cannot open $output\n";

#
# printout the date and version lines in two parts so that RCS will not
# remove the \ escape on the $.
#
print GP2P "!date: \$Date: ";
print GP2P " \$\n";
print GP2P "!version: \$Revision: ";
print GP2P " \$ \n!\n";
print GP2P "!Sacchharomyces Genome Database\n";
print GP2P "!SGDID mapped to SwissPROT/TrEMBL Accessions\n";
print GP2P "!\n";

while (<>) {
    chop;

    my $line = uc($_);

    if ($line =~ /^[SL]/) {
#	my ($sgdid) = split(/\t/, $line, 1);
	my $sgdid = $_;

	if ($swiss{$sgdid}) {
#
# for now include SGD: in the first column
	    print GP2P "SGD:" . $sgdid . "\tSWISS-PROT:" . $swiss{$sgdid} . "\n";
	}
    }
}

close (GP2P);

exit 1;