Script to test and verify fasta files are ready for InParanoid

svn path=/; revision=29

diff --git a/interactome_scripts/fasta_verify.pl b/interactome_scripts/fasta_verify.pl
new file mode 100755 (executable)
index 0000000..6f12a38
--- /dev/null
+++ b/interactome_scripts/fasta_verify.pl
@@ -0,0 +1,52 @@
+#!/usr/bin/perl
+
+use strict;
+use warnings;
+
+
+if($#ARGV != 0) {
+       print "usage: fasta_verify.pl fast_file\n";
+       exit;
+}
+
+my $in_file = $ARGV[0];
+
+open(in_file, "$in_file");
+
+my $prev_line_is_header = 0;
+
+my $counter = 0;
+
+while(<in_file>) {
+       my $line = $_;
+       chomp $line;
+       $counter++;
+       
+       # detect if two lines in a row are headers, if they are, print error and give line #
+       my $curr_line_is_header = 0;
+       if($line =~ /^>/) {
+               $curr_line_is_header = 1;
+       }
+       
+       if($curr_line_is_header == 1 && $prev_line_is_header == 1) {
+               print "Error: Header found with no sequence data on line $counter in file $in_file\n";
+       }
+       
+       # detect if line after header is blank
+       if($line !~ /^[a-zA-Z*]/ && $prev_line_is_header == 1) {
+               print "Error: Blank line found on line $counter in file $in_file\n";
+       }
+       
+       # detect if header line has tabs in it
+       if($curr_line_is_header ==1 && $line =~ /\t/) {
+               print "Error: Tab character in header on line $counter in file $in_file\n";
+               print "Reccommend using sed 's/\\t/\|/g' to replace tabs with pipe symbols\n";
+       }
+
+       # set prev_line_header info for next time through loop
+       if($curr_line_is_header) {
+               $prev_line_is_header = 1;
+       }else{
+               $prev_line_is_header = 0;
+       }
+}