[Bio] / FortyEight / rp_make_interpro_download_attributes.pl Repository:
ViewVC logotype

View of /FortyEight/rp_make_interpro_download_attributes.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.1 - (download) (as text) (annotate)
Wed Feb 27 23:27:54 2008 UTC (11 years, 9 months ago) by mkubal
Branch: MAIN
CVS Tags: mgrast_dev_08112011, rast_rel_2009_05_18, mgrast_dev_08022011, rast_rel_2014_0912, rast_rel_2008_06_18, rast_rel_2008_06_16, mgrast_dev_05262011, rast_rel_2008_12_18, mgrast_dev_04082011, rast_rel_2008_07_21, rast_rel_2010_0928, rast_2008_0924, mgrast_version_3_2, mgrast_dev_12152011, rast_rel_2008_04_23, mgrast_dev_06072011, rast_rel_2008_09_30, rast_rel_2009_0925, rast_rel_2010_0526, rast_rel_2014_0729, mgrast_dev_02212011, rast_rel_2010_1206, mgrast_release_3_0, mgrast_dev_03252011, rast_rel_2010_0118, mgrast_rel_2008_0924, mgrast_rel_2008_1110_v2, rast_rel_2009_02_05, rast_rel_2011_0119, mgrast_rel_2008_0625, mgrast_release_3_0_4, mgrast_release_3_0_2, mgrast_release_3_0_3, mgrast_release_3_0_1, mgrast_dev_03312011, mgrast_release_3_1_2, mgrast_release_3_1_1, mgrast_release_3_1_0, mgrast_dev_04132011, rast_rel_2008_10_09, mgrast_dev_04012011, rast_release_2008_09_29, mgrast_rel_2008_0806, mgrast_rel_2008_0923, mgrast_rel_2008_0919, rast_rel_2009_07_09, rast_rel_2010_0827, mgrast_rel_2008_1110, myrast_33, rast_rel_2011_0928, rast_rel_2008_09_29, mgrast_rel_2008_0917, rast_rel_2008_10_29, mgrast_dev_04052011, mgrast_dev_02222011, rast_rel_2009_03_26, mgrast_dev_10262011, rast_rel_2008_11_24, rast_rel_2008_08_07, HEAD
new

my $jobdir = shift(@ARGV);
my $genome = shift(@ARGV);

my $attribute_file =  "$jobdir/rp/$genome/attributes/interpro_download_attributes.txt";
my $crc64_mapping_file = "$jobdir/rp/$genome/crc64_for_pegs_of_$genome.txt";

open(RESULTS,">$attribute_file");

my @files = ("uniparc_match_1.dump","uniparc_match_2.dump","uniparc_match_3.dump","uniparc_match_4.dump","uniparc_match_5.dump","uniparc_match_6.dump","uniparc_match_7.dump");

my %id_crc;

open(IN,$crc64_mapping_file);
while($_ = <IN>){
    chomp($_);
    my ($id,$crc) = split("\t",$_);
    $id_crc{$crc} = $id;
}
close(IN);

my $interpro_download_dir = "/vol/seed-attributes/Interpro_Download_Dump";
foreach my $file (@files){
    open(IN,"$interpro_download_dir/$file");
    my $crc;
    my $db;
    my $db_id;

    my $id;
    my $bitscore;
    my $evalue;
    my $length;
    my $start;
    my $end;
    my $record = 0;

    my $interpro_id;
    
    while ($_ = <IN>){
	if($record){
	    if($_ =~ /match id=\"(\w+\d+)\".*dbname=\"(\w+)\"/){
		$db_id = $1;
		$db = $2;
	    }
	    
	    if($_ =~ /ipr id=\"(\w+\d+)\"/){
                $interpro_id = $1;
	    }
	    
	    if($_ =~ /lcn start=\"(\d+)\"\s+end=\"(\d+)\"\s+score=\"(.*)\"/){
		$start = $1;
		$end = $2;
		$bitscore = $3;
		my $db_size = 679928271;
		
		if($db){
		    print RESULTS "$id\t$db"."::$db_id"."_interpro_download\t$bitscore;$start-$end\n";
		    print RESULTS "$id\tIPR::$interpro_id"."_interpro_download\t$bitscore;$start-$end\n";
		}
	    }
	}
	
	if($_ =~/length=\"(\d+)\"\s+crc64=\"(.*)\"/){
	    $length = $1;           
            $crc = $2;
	    $record = 0;
	}
	if($id_crc{$crc}){
	    $id = $id_crc{$crc};
	    $record = 1;
	}
    }
    close(IN);
}

print RESULTS "$genome\tInterpro_Domain_Download\tRelease 15.0\n";
close(RESULTS);

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3