[Bio] / FigKernelScripts / package_sims.pl Repository:
ViewVC logotype

View of /FigKernelScripts/package_sims.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.2 - (download) (as text) (annotate)
Mon Dec 5 18:56:37 2005 UTC (13 years, 11 months ago) by olson
Branch: MAIN
CVS Tags: mgrast_dev_08112011, rast_rel_2009_05_18, mgrast_dev_08022011, rast_rel_2014_0912, rast_rel_2008_06_18, myrast_rel40, 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, caBIG-05Apr06-00, 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, caBIG-13Feb06-00, rast_rel_2009_03_26, mgrast_dev_10262011, rast_rel_2008_11_24, rast_rel_2008_08_07, HEAD
Changes since 1.1: +17 -0 lines
Add license words.

# -*- perl -*-
#
# Copyright (c) 2003-2006 University of Chicago and Fellowship
# for Interpretations of Genomes. All Rights Reserved.
#
# This file is part of the SEED Toolkit.
# 
# The SEED Toolkit is free software. You can redistribute
# it and/or modify it under the terms of the SEED Toolkit
# Public License. 
#
# You should have received a copy of the SEED Toolkit Public License
# along with this program; if not write to the University of Chicago
# at info@ci.uchicago.edu or the Fellowship for Interpretation of
# Genomes at veronika@thefig.info or download a copy from
# http://www.theseed.org/LICENSE.TXT.
#

use Carp 'cluck';
use Data::Dumper;
use P2Pupdate;
use strict;
use Sim;

use FIG;
my $fig = new FIG;
my $dbh = $fig->db_handle();

my %file_map;

# usage: package_sims SimsPackage genome

sub usage
{
    die "Usage: $0  SimsPackage genome\n";
}

my $package = shift @ARGV;
my $genome = shift @ARGV;

$genome =~ /^\d+\.\d+$/ or die "Invalid genome $genome\n";

open(PKG, ">$package") or die "Could not open package file $package for writing: $!\n";

my $warning_count = 0;
my $max_warnings = 20;

#
# Gather pegs.
#

my @pegs = sort { &FIG::by_fig_id($a, $b) } $fig->pegs_of($genome);

my %pegs;

map { $pegs{$_}++ } @pegs;

my %pegs_left = %pegs;

my $dbi = $dbh->{_dbh};

my $sim_seek_stmt = $dbi->prepare(qq(SELECT filen, seek, len
				     FROM sim_seeks
				     where id  = ?));

for my $peg (@pegs)
{
    #
    # Determine if this id is the representative ID in the peg.synonyms.
    #
    
    my @mapped = $fig->mapped_prot_ids($peg);

    my $rep = $mapped[0]->[0];

    if (not defined($rep))
    {
	print "Couldn't map $peg\n";
	next;
    }

    #
    # Retrieve sims based on $rep
    #

    $sim_seek_stmt->execute($rep);
    my $sim_entries = $sim_seek_stmt->fetchall_arrayref();

    my $delta = 0;
    my @entry;
    if ($rep ne $peg)
    {
	#
	# Determine the offsets we need to use in adjusting sim output.
	#
	

	@entry = grep { $_->[0] eq $peg } @mapped;
	if (not (@entry == 1 and defined($entry[0]->[1])))
	{
	    die "bad entry from mapped list\n";
	}

	$delta = $mapped[0]->[1] - $entry[0]->[1];
    }
    
    #
    # Read the sims, and map names/lengths if necessary.
    #

    if (not $sim_entries or @$sim_entries == 0)
    {
	print STDERR "No sims found for peg=$peg rep=$rep ", Dumper(\@mapped, \@entry);
	next;
    }

    for my $ent (@$sim_entries)
    {
	my($filen, $seek, $len) = @$ent;

	my $file = map_file($filen);
	my $fh = $fig->openF($file);
	$fh or die "Cannot opensims file for $file $filen\n";
	
	my $block = &FIG::read_block($fh, $seek, $len);
	    
	#
	# Walk thru the sims in the block.
	#
	# If this peg isn't the representative synonym, adjust the id1
	# name to our peg and shift the match position in id1. (since
	# the tails are the same, but the positions are reported with
	# respect to the head).
	#

	for my $line (@$block)
	{
	    my $fixed = ($line =~ s/\t\t/\t/g);

	    my $sim = Sim->new_from_line($line);
	    
	    if (!$sim->validate)
	    {
		$warning_count++;

		cluck "Invalid similarity for $sim, skipping" if $warning_count < $max_warnings;
		next;
	    }
	    
	    if ($fixed)
	    {
		$warning_count++;
		cluck "Fixed double-tab in similarity $sim" if $warning_count < $max_warnings;
	    }

	    if ($rep ne $peg)
	    {
		$sim->[0] = $peg;
		$sim->[6] -= $delta;
		$sim->[7] -= $delta;
	    }

	    print PKG $sim->as_line;
	}
    }
}

print "Package complete with $warning_count warnings\n";
close(PKG);

sub map_file
{
    my($filen) = @_;

    my $mapped = $file_map{$filen};
    if (not defined($mapped))
    {
	$mapped = $fig->N2file($filen);
	$file_map{$filen} = $mapped;
    }
    return $mapped;
}

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3