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

View of /FigKernelScripts/add_rbs_data_for_start_predictions.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.3 - (download) (as text) (annotate)
Fri Nov 30 21:53:05 2007 UTC (11 years, 11 months ago) by gdpusch
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, 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
Changes since 1.2: +18 -3 lines
Added FIGV support. -- /gdp

# -*- 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.
########################################################################

$usage = "add_rbs_data_for_start_predictions < stream_of_ORF_objects > with_RBS_scores";

$trouble   = 0;
while (@ARGV)
{
    if ($ARGV[0] =~ m/-h(elp)?/) {
	die "\n\tusage: $usage\n\n";
    }
    else {
	warn "Invalid argument $ARGV[0]\n";
    }
    shift @ARGV;
}
die "\nusage: $usage\n\n" if $trouble;


$/ = "///\n";
while (defined($obj = <STDIN>))
{
    chomp $obj;
    if ($obj =~ /^(.*?\n)\/\/\n(.*)/s)
    {
	($hdr,$start_data) = ($1,$2);
	print $hdr;
	
	$kv = &load_kv($hdr);
	$seq = $kv->{'SEQ'};
	$pre = $kv->{'PREFIX'};
	$ln_pre   = length($pre);
	$with_pre = $pre . $seq;

	@poss_starts = split(/\n\/\/\n/,$start_data);
	foreach $poss_start (@poss_starts)
	{
	    if ($poss_start =~ /START=(\d+)/s)
	    {
		$start = $1;
		($sd_beg,$sd_end) = &best_rbs($with_pre,$start+$ln_pre);
		$sd_len = defined($sd_beg) ? $sd_end+1-$sd_beg : 0;
		# if (defined($sd_beg)) { print "RMB: BESTRBS=", substr($with_pre,$sd_beg,($sd_len)), "\n"; }
		print "//\n";
                print "$poss_start\n";
                print "RBS_LEN=$sd_len\n";
                ## compute distance to left of this start
                print "RBS_POS=", defined($sd_beg) ? -($ln_pre+$start-$sd_beg) : 0, "\n";
	    }
	}
	if (@poss_starts) { print "//\n"; }
	print "///\n";
    }
}

sub best_rbs {
    my($seq,$start) = @_;
    my($best,$i,$str,$len_sd,$sd_beg,$sd_end);

    $best = 0;
    $best_rbs = '';
    for ($i = ($start >= 16) ? 16 : $start,$len_sd=0; ($i >= 6); $i--)
    {
	$str = uc substr($seq,($start - $i),8);
	$len_sd = &shine_dalgarno($str,$i-6);
	# print "CHECKING $start $i ", $start-$i, " $best $len_sd\n";
	if ((($i-$len_sd) <= 9) &&  (($len_sd > $best) || (($len_sd == $best) && (($i-$len_sd) >= 2))))
	{
	    $best = $len_sd;
	    $sd_beg = $start - $i;
	    $sd_end = $sd_beg + ($len_sd-1);
	    $best_rbs = substr($str,0,$len_sd);
	}
    }

    if ($best)
    {
	# print "RMB: BESTRBS=$best_rbs\n";
	return ($sd_beg,$sd_end);
    }
    else
    {
	return (undef,undef);
    }
}


sub shine_dalgarno {
    my($seq,$toright) = @_;
    my($sub);
# RGGRGGTGAT = Shine-Dalgarno (match any contiguous 4)

    $sub   = substr($seq,0,4);

    if    ($sub =~ /^[AG]GG[AG]/) { return 4 + &extend($seq,4,"GGTGAT",$toright); }
    elsif ($sub =~ /^GG[AG]G/)    { return 4 + &extend($seq,4,"GTGAT",$toright); }
    elsif ($sub =~ /^G[AG]GG/)    { return 4 + &extend($seq,4,"TGAT",$toright); }
    elsif ($sub =~ /^[AG]GGT/)    { return 4 + &extend($seq,4,"GAT",$toright); }
    elsif ($sub =~ /^GGTG/)       { return 4 + &extend($seq,4,"AT",$toright); }
    elsif ($sub =~ /^GTGA/)       { return 4 + &extend($seq,4,"T",$toright); }
    elsif ($sub =~ /^TGAT/)       { return 4; }
    return 0;
}

sub extend {
    my($seq,$start,$against,$toright) = @_;
    my($ln,$i,$c1,$c2);
    my($max) = $toright;

#    print STDERR "extending $toright\n";
    $ln = length($against);
    $i  = 0;
    $c1 = substr($seq,($start + $i),1);
    $c2 = substr($against,$i,1);
    while ($max-- && $ln && ($c1 eq $c2))
    {
	$i++; $ln--;
	$c1 = substr($seq,($start + $i),1);
	$c2 = substr($against,$i,1);
    }
    return $i;
}

sub load_kv {
    my($obj) = @_;
    my($kv,$kvH);

    $kvH = {};
    foreach $kv (split(/\n/,$obj))
    {
	if ($kv =~ /^([^=]+)=(.*)$/)
	{
	    $kvH->{$1} = $2;
	}
    }
    return $kvH;
}

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3