Added script that parses NCBI taxonomy.dat file and creates slim OWL file, to be...

svn path=/; revision=629

diff --git a/interactome_scripts/planteome_ncbi_taxon_slimmer.pl b/interactome_scripts/planteome_ncbi_taxon_slimmer.pl
new file mode 100755 (executable)
index 0000000..a74e3ee
--- /dev/null
+++ b/interactome_scripts/planteome_ncbi_taxon_slimmer.pl
@@ -0,0 +1,99 @@
+#!/usr/bin/perl
+
+###############################################################
+#  Justin Elser  (elserj@science.oregonstate.edu)             #
+#              Parses the assoc files for any taxon ids              #
+#              used in planteome assoc files.  Uses that to          #
+#              generate small (slim) subset of the ncbi taxonomy     #
+#              dat file.  From there, uses the owltools                          #
+#              ncbi2owl.jar program to generate OWL file             #
+#                                                             #
+###############################################################
+
+
+use strict;
+use warnings;
+
+# check for arguments and explain usage
+if ($#ARGV !=2) {
+       print "usage: planteome_ncbi_taxon_slimmer.pl taxonomy.dat_file taxonomy_slim_file output_OWL_file\n";
+       exit;
+}
+
+my $taxon_file = $ARGV[0];
+my $slim_file = $ARGV[1];
+my $output_file = $ARGV[2];
+
+my $ncbi_owltools_path = "/home/justin/owltools/OWLTools-NCBI/bin/ncbi2owl.jar";
+
+# open the assoc files to be parsed, assume this is all sqltable* files in directory
+
+my @files = glob("*.assoc");
+
+my %taxon_hash;
+# taxon id 1 is required, I think
+$taxon_hash{1} = 1;
+
+# get the relevant taxon ids
+foreach my $input_file (@files) {
+       open(INFILE, "$input_file") || die "Error: file '$input_file' can not be opened\n";
+       while(<INFILE>){
+                       my $line = $_;
+                       chomp $line;
+                       
+                       if($line =~ /taxon:(\d+)/) {
+                                       my $taxon = $1;
+                                       if(!defined($taxon_hash{$taxon})){
+                                                       $taxon_hash{$taxon} = $taxon;
+                                       }
+                       }
+       }
+       close(INFILE);
+}
+
+
+
+# Read in the full taxon file from ncbi and output only the entries that have matching taxon ids
+open(TAXONFILE, "$taxon_file") || die "Error: file '$taxon_file' can not be opened\n";
+open(TAXONSLIM, ">$slim_file") || die "Error: file '$slim_file' can not be opened for writing\n";
+my $keep_section = 0;
+while(<TAXONFILE>){
+               my $line = $_;
+               chomp $line;
+               
+               if($line =~ /^ID\s+:\s(\d+)/) {
+                               if(defined($taxon_hash{$1})) {
+                                               print TAXONSLIM "$line\n";
+                                               $keep_section = 1;
+                                               next;
+                               }
+               }
+               
+               if($keep_section == 1) {
+                               print TAXONSLIM "$line\n";
+               }
+               
+               if($line =~ /^\/\//) {
+                               $keep_section = 0;
+               }
+}
+                               
+                                               
+system("java -Xmx6G -jar $ncbi_owltools_path $slim_file $output_file");
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+                                               
+