2
|
1 #!/usr/bin/perl
|
|
2
|
|
3 # velvet_optimiser_vlsci.pl
|
|
4 #
|
|
5 # Copyright 2012 Simon <simon@Hyperion>
|
|
6 #
|
|
7 # This program is free software; you can redistribute it and/or modify
|
|
8 # it under the terms of the GNU General Public License as published by
|
|
9 # the Free Software Foundation; either version 2 of the License, or
|
|
10 # (at your option) any later version.
|
|
11 #
|
|
12 # This program is distributed in the hope that it will be useful,
|
|
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
15 # GNU General Public License for more details.
|
|
16 #
|
|
17 # You should have received a copy of the GNU General Public License
|
|
18 # along with this program; if not, write to the Free Software
|
|
19 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
|
|
20 # MA 02110-1301, USA.
|
|
21 #
|
|
22 #
|
|
23
|
|
24 use strict;
|
|
25 use warnings;
|
|
26 use File::Copy;
|
|
27
|
|
28 my @stuff = @ARGV;
|
|
29 my %counts;
|
|
30 my %shortMPs;
|
|
31 my @reads;
|
|
32 my $threadsToUse = 2; #According to Enis, this is the default number of threads that multithreaded programs use..
|
|
33
|
|
34 #TO DO! get velvet compile parameters!
|
|
35
|
|
36 foreach my $param (@stuff){
|
|
37 if($param =~ m/galaxy\.datatypes/){
|
|
38 $param =~m/datatypes\..+\.(.+)\s+object/;
|
|
39 my $type = lc($1);
|
|
40 print "$type\n";
|
|
41 }
|
|
42 else {
|
|
43 print $param . "\n";
|
|
44 }
|
|
45 }
|
|
46
|
|
47 #try and split some of it up..
|
|
48 #kmer stuff
|
|
49 my $sk = shift @stuff;
|
|
50 my $ek = shift @stuff;
|
|
51 my $xk = shift @stuff;
|
|
52
|
|
53 #TO DO! do checks etc on this stuff here!
|
|
54
|
|
55 #get the read file stuff
|
|
56 my $i = 0;
|
|
57
|
|
58 while(!($stuff[0] =~ m/other:/)){
|
|
59 my $index = shift @stuff;
|
|
60 my $read_type = shift @stuff;
|
|
61 my $sep = shift @stuff;
|
|
62 my $file_type = shift @stuff;
|
|
63 $file_type =~m/datatypes\..+\.(.+)\s+object/;
|
|
64 $file_type = lc($1);
|
|
65 $file_type = "fastq" if $file_type =~ m/fastq/i;
|
|
66 $counts{$read_type} ++;
|
|
67 my $ffile = shift @stuff;
|
|
68 my $rfile = shift @stuff if $sep eq "True";
|
|
69 my $sMP = shift @stuff;
|
|
70 $reads[$i] = "-$read_type";
|
|
71 $reads[$i] .= $counts{$read_type} if $counts{$read_type} > 1;
|
|
72 if ($sMP eq "shortMP_lib"){
|
|
73 my $temp = "-shortMatePaired";
|
|
74 $temp .= $counts{$read_type} if $counts{$read_type} > 1;
|
|
75 $shortMPs{$temp} = 1;
|
|
76 }
|
|
77 $reads[$i] .= " -$file_type";
|
|
78 $reads[$i] .= " -separate" if $sep eq "True";
|
|
79 $reads[$i] .= " $ffile";
|
|
80 $reads[$i] .= " $rfile" if $sep eq "True";
|
|
81 $i ++;
|
|
82 }
|
|
83
|
|
84 #get the other stuff
|
|
85 shift @stuff if $stuff[0] =~ m/other:/;
|
|
86 my $amos = shift @stuff;
|
|
87 my $verbose = shift @stuff;
|
|
88
|
|
89 #get the advanced stuff if it exists!
|
|
90 my ($oFK, $oFC, $vgO, $minC, $maxC);
|
|
91 if ($stuff[0] =~ m/advanced:/) {
|
|
92 shift @stuff;
|
|
93 $oFK = shift @stuff;
|
|
94 $oFC = shift @stuff;
|
|
95 $vgO = shift @stuff;
|
|
96 $minC = shift @stuff;
|
|
97 $maxC = shift @stuff;
|
|
98 }
|
|
99
|
|
100 #get the output file names!
|
|
101 my $contigs_outfile = shift @stuff;
|
|
102 my $stats_outfile = shift @stuff;
|
|
103 my $afg_outfile = shift @stuff if $amos eq "amos";
|
|
104 my $logfile = shift @stuff;
|
|
105 my $stderr = shift @stuff;
|
|
106 my $lgraph = shift @stuff;
|
|
107
|
|
108 #build the command line...
|
|
109
|
|
110 my $cmd = "VelvetOptimiser.pl -d temp_vgo_files -s $sk -e $ek -x $xk";
|
|
111 $cmd .= " -a" if($amos eq "amos");
|
|
112 $cmd .= " -v" if($verbose eq "verbose");
|
|
113
|
|
114 $cmd .= " -f '";
|
|
115 foreach my $line (@reads){
|
|
116 $cmd .= " $line";
|
|
117 }
|
|
118 $cmd .= "'";
|
|
119
|
|
120 my $vgopt = $vgO;
|
|
121 foreach my $key(keys %shortMPs){
|
|
122 $vgopt .= " $key" if $key;
|
|
123 }
|
|
124
|
|
125 $cmd .= " -o '$vgopt'" if $vgopt;
|
|
126
|
|
127 $cmd .= " -k $oFK" if $oFK;
|
|
128 $cmd .= " -c $oFC" if $oFC;
|
|
129 $cmd .= " -m $minC" if $minC;
|
|
130 $cmd .= " -z $maxC" if $maxC;
|
|
131 $cmd .= " -t $threadsToUse";
|
|
132
|
|
133 $cmd .= " 2> $stderr";
|
|
134
|
|
135 print "\n$cmd\n";
|
|
136
|
|
137 print "about to run the command!";
|
|
138
|
|
139 if(system($cmd) == 0) {
|
|
140 #copy the files to the new places and delete the directory..
|
|
141 print "Copying output\n";
|
|
142 copy("temp_vgo_files/contigs.fa", $contigs_outfile);
|
|
143 copy("temp_vgo_files/stats.txt", $stats_outfile);
|
|
144 copy(glob("temp_vgo_files/*Logfile.txt"), $logfile);
|
|
145 copy("temp_vgo_files/velvet_asm.afg", $afg_outfile) if $amos eq "amos";
|
|
146 copy("temp_vgo_files/LastGraph", $lgraph);
|
|
147 system("rm -rf temp_vgo_files") == 0 or die "Couldn't delete temporary directory. $!";
|
|
148 }
|
|
149 else {
|
|
150 print "There was a velvet optimiser error\n";
|
|
151 exit(1);
|
|
152 }
|
|
153
|
|
154 exit(0);
|