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

View of /FigKernelScripts/reformat_timelogic_sims.pl

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.2 - (download) (as text) (annotate)
Wed Sep 5 21:12:21 2007 UTC (12 years, 2 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, 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 1.1: +1 -1 lines
rollup of changes for anno clearinghouse, RAST->SEED import.

#
# Reformat sims that come from the Decypher / timelogic hardware processor board.
#
# We assume they come from the ANL installation using the tera-blastp-tabular
# format.
#
# The files look something like this:
#
# QUERYLOCUS      TARGETLOCUS     PERCENTTARGET   ALIGNMENTLENGTH MATCHES GAPS    SIMILARITIES    QUERYSTART      QUERYEND       TARGETSTART     TARGETEND       SCORE   SIGNIFICANCE    QUERYLENGTH     TARGETLENGTH
# xxx04279584     xxx04279584     93      1575    1492    0       1492    18      1592    18      1592    2967.56 0.000000       1592    1592
#
#
# This script reads the first line to determine the fields that are in this dataset.
# The %field_map table is consulted to generate a mapping vector used to reorder the
# columns to match the usual blastall -m8 output.
#

use strict;
use Data::Dumper;

my %field_map = (QUERYLOCUS => 0,
		 TARGETLOCUS => 1,
		 PERCENTALIGNMENT => 2,
		 ALIGNMENTLENGTH => 3,
		 MATCHES => undef,
		 GAPS => 5,
		 SIMILARITIES => undef,
		 QUERYSTART => 6,
		 QUERYEND => 7,
		 TARGETSTART => 8,
		 TARGETEND => 9,
		 SCORE => 11,
		 SIGNIFICANCE => 10,
		 QUERYLENGTH => undef,
		 TARGETLENGTH => undef,
		);

#
# Dest_field_map maps from destination column to fieldname.
#
my %dest_field_map = map { $field_map{$_} => $_ } keys %field_map;

$_ = <>;

if (!/QUERY/)
{
    die "Invalid timelogic sims input (no QUERY found on first line\n";
}

chomp;
s/^\s*//;
my @fields = split(/\s+/, $_);
my %fields;
my $i = 0;
map { $fields{$_} = $i++ } @fields;

my @shuffle;
for my $dest_col (keys %dest_field_map)
{
    $shuffle[$dest_col] = $fields{$dest_field_map{$dest_col}};
}
@shuffle = map { defined($_) ? $_ : 999 } @shuffle;

while (<>)
{
    chomp;
    
    my @in = split(/\s+/);

    my @out;

    @out = map { defined($_) ? $_ : -1 } @in[@shuffle];

    print join("\t", @out), "\n";
}

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3