GEN220 HW5 BioPerl

Bio.DB.Fasta

#!/usr/bin/perl -w
use strict;
use warnings;

use Bio::DB::Fasta;
use Bio::Seq;
use Bio::SeqIO;

my $file = 'sacharomyces_cerevisiae_S288C.fa';
my $db = Bio::DB::Fasta->new($file);

my $piece = $db->seq('chrI',54584 => 54913);
print "Sequence from ChromosomeI 54584..54913.\n", "$piece\n";

my $seq = Bio::Seq->new(-seq =>$piece);
print "Amino acid sequence.\n", $seq->translate->seq, "\n";

Bio.SeqIO

#!/usr/bin/perl
use strict;
use warnings;
use Bio::SeqIO;

my $file = 'seqs.gbk';
my $in  = Bio::SeqIO->new(-file => $file ,
                           -format => 'genbank');

my ($count);
while ( my $seq = $in->next_seq) {
 $count++;

print $seq->display_id, ":\n";
 my $cds = 0;
  for my $feature ( $seq->get_SeqFeatures ) {
  if ( $feature->primary_tag eq 'CDS' ) {
   $cds ++;
   my $length = $feature->end - $feature->start;
   print "CDS Identified! Length is $length\n";
}
}
print "Number of CDS features is $cds\n";
}

printf $file, $count;

Blast Report

#!/usr/bin/perl -w
use strict;
use Bio::SearchIO;
my $cutoff = '0.001';
my $file = 'basidio.BLASTP';
my $in = Bio::SearchIO->new(-format => 'blast',
                                -file => $file);

while( my $r = $in->next_result ) {
print "Query is: ", $r->query_name, " ",
$r->query_description," ",$r->query_length," aa\n";
print " Matrix was ", $r->get_parameter('matrix'), "\n";
while( my $h = $r->next_hit ) {
last if $h->significance > $cutoff;
print "Hit is ", $h->name, "\n";
while( my $hsp = $h->next_hsp ) {
print " HSP Len is ", $hsp->length('total'), " ",
" E-value is ", $hsp->evalue, " Bit score ",
$hsp->score, " \n",
" Query loc: ",$hsp->query->start, " ",
$hsp->query->end," ",
" Sbject loc: ",$hsp->hit->start, " ",
$hsp->hit->end,"\n";
}
}
}

last one is just a copy of the slides from the class and doesn't print out the tab delimited report as requested.