How to import fasta file by using perl DBI.?

use DBI;
use strict;
use warning;

my $textfile = "fasta.txt";
open (TEXT,"$textfile") or die "Could not open:$!";

my $dbh = DBI->connect("dbi:mysql:database=file",user,password,{RaiseError=>1}) or die "Could not connect $DBI:errstr";

while(<TEXT>){
    $first = <TEXT>;
    @second = <TEXT>;
    $dbh->do("INSERT INTO open values (?,?)", undef, $first,@second);
    }

$dbh->disconnect;
close TEXT;

This is My Perl code. It is not insert data in database. I created database and table. database name is file and table name is open, the table contain two column one is header and another one is sequence.

My expectation. I will create database have two column, one column contain the first line of fasta file and second column contain their sequence.

Header                                                                                        Sequence
>sp|P00547|KHSE_ECOLI Homoserine kinase OS=Escherichia coli (strain K12) GN=thrB PE=1 SV=2    MVKVYAPASSANMSVGFDVLGAAVTPVDGALLGDVVTVEAAETFSLNNLGRFADKLPSEP
                                                                                              RENIVYQCWERFCQELGKQIPVAMTLEKNMPIGSGLGSSACSVVAALMAMNEHCGKPLND

Please help me..

database dbi perl

• 3,457 views • link

updated 4.2 years ago by Ram

• written 11.4 years ago by Krithi.S •

1 answer

First your code should not run: you 'use strict' and declare variables without 'my'. Second, you're not using placeholders correctly. Each ? has to map to a scalar so (?,?) requires two scalars and it looks like you have a scalar ($first) and a list (@second). However, using $dbh->do() in this way is inefficient because you create and destroy a statement handle at each iteration. A more efficient way to use placeholders is with prepare:

my $query = qq(INSERT INTO open values (?,?));
my $sth = $dbh->prepare($query);
while(<TEXT>){
    my $first = <TEXT>;
    my $second = <TEXT>; # get the next line of text only
    $sth->execute($first,$second);
}

Finally, you also need to parse your fasta file properly.

• 0 views • link

updated 4.2 years ago by Ram

• written 11.4 years ago by Jean-Karim Heriche

@Jean-Karim Heriche, thanking you for your nice answer. But I am not able to getting Header in my output in one column? Only my output is have a sequence, but its to split into three rows. I want to expect in that header column only the first line and that sequence column contain all sequence in fasta file. Please help me.

• 0 views • link

updated 4.2 years ago by Ram

• written 11.4 years ago by Krithi.S •

This is what I meant by parsing your fasta file correctly. I suggest you use BioPerl's Bio::SeqIO for this. However, if you need an alternative, here is a function that does the job:

=head2 get_fasta_record

 Arg: opened filehandle
 Description: Reads next record from a fasta file
 Returntype: arrayref to (header, sequence)

=cut

sub get_fasta_record {
  my $fh = shift;
  local $/ = ">";
  my ($header,$sequence);
  while(my $line =<$fh>) {
    chomp($line);
    $line=~s/\R/\n/g; # replace any line break characters with \n
    next if ($line=~/^\s*$/);
    my $newline_pos = index($line,"\n");
    $header = substr($line,0,$newline_pos);
    $sequence = substr($line,$newline_pos+1);
    $sequence =~s/\n//g;
    $sequence =~s/\s+//g;
    last;
  }
  if (defined($header)) {
    return [$header,$sequence];
  }
  else { return undef; }
}

Use like this:

open my $fh,"<",$ARGV[0] or die "\nERROR: Can't read file $ARGV[0]: $!";
while (my ($header,$sequence) = @{$seq_reader->get_fasta_record($fh)}) {
  # Do something here
}
close $fh;

• 0 views • link

updated 4.2 years ago by Ram

• written 11.4 years ago by Jean-Karim Heriche