Parse a genbank file using regular expressions

Nishi Singh — Tue, 10 May 2016 11:56:26 -0500

#! /usr/local/bin/perl -w

$genbank = "genbank_file.txt";

open (GENBANK, $genbank) || die "cannot open $gb_report for reading: $!";

# Flag for multiline translation; 1 means translation "in progress"  
$trans = 0;

while ()
{
   if (/(LOCUS\s*)(\w*)(.*)/) { 
       print "Locus: $2\n"; 
   }
   elsif (/(VERSION.*GI:)(\d*)/) { 
      print "GI: $2\n"; 
   }
   elsif (/(DEFINITION\s*)(.*)(\.)/) {
      print "Sequence name: $2\n";
   }
   elsif (/(ORGANISM\s*)(.*)/) {
      print "Organism: $2\n";
   }
   elsif(/(gene)(\s*)(\d*)(\.\.)(\d*)/) {
      print "Gene length: $5\n";
   }
   elsif (/(CDS\s*)(\d*)(\.\.)(\d*)/)  {
   # ex: CDS             357..1541
      $cds_start = $2;
      $cds_end = $4;
      print "CDS: $cds_start - $cds_end\n";
   }
   elsif (/(\/translation=")(.*)/)  {  # protein product begins
      print "Translation: ";
      $protein = $2;
      $trans = 1;
   }
   elsif ($trans)  {   # translation still going on
      if (!/"/)  {  # no terminal quote; translation continues
         $protein .= $_;
      }
      elsif (/(.*)(")/)  {  # terminal quote; end of translation
         $protein .= $1;
         $protein =~ s/\s*//g;
         print "$protein\n";
         $trans = 0;
      }
      else  {
         print "Problems: end of translation product not found.\n";
      }
   }
   else  {
      # Skip this data
   }
}

BOL: Parse a genbank file using regular expressions

Parse a genbank file using regular expressions