THIS FACILITY IS PROVIDED BY NCBI ITSELF..HOPE THIS WILL HELP

http://www.ncbi.nlm.nih.gov/sites/batchentrez

The simplest way is to use E-Utilities, which are already ready and written in perl. You can pass the same type of complex queries one can use in the web interface of Entrez (including ranges). The problem in your case is that you can't use accn of protein to get its gene/mrna directly. Given that you have GenPept, you could use it to retrieve the DBSOURCE locus/accession of the gene and feed it into Efetch.

I think you could genpept accn to retrieve genes in NCBI.

Edit:

This solution works for finished genomes with annotated genes and transcripts. For drafts the situation is much more delicate and depends heavily on the annotation quality.

Try BioMart

/#!/usr/bin/perl
    use strict;
    use warnings;
    use Bio::DB::GenBank;
    use Bio::SeqIO;
    # get command-line arguments, or die with a usage statement
    my $usage = "RetriveDataGenBank.pl Accessions_infile Fasta_outfilen";
    my $infile = shift or die $usage;
    my $outfile = shift or die $usage;
    my $outfileformat = 'Fasta';
    my $gb = Bio::DB::GenBank->;new(-retrievaltype => 'tempfile' , -format => 'Fasta');
    my @Acc_Numbers;
    open (IN,$infile)||die;
    my $line;
    while ($line =[?]) {
        chomp $line;
        push (@Acc_Numbers, "'".$line."'")
    }
    my $seq_in = $gb->get_Stream_by_acc(@Acc_Numbers);
    my $seq_out = Bio::SeqIO->new('-file' => ">$outfile",'-format' => $outfileformat);
    # write each entry in the input file to the output file
    while (my $inseq = $seq_in->next_seq) {
        $seq_out->write_seq($inseq);
    }
    exit

your INPUT - list of GI's
your OUTPUT - Fasta sequences.fasta

Ilia

Use eutils as follow: 1. first search each protein against the "gene" database to get genes ID's of each protein. - use esearch 2. the search these genes in the "nucleotide" database. - esearch again. 3. use efetch to download the nucleotide FASTA sequences. see also: http://www.ncbi.nlm.nih.gov/books/NBK1058/

Hello everyone,
I'd like to show you the code that I trying with:

#/usr/bin/perl
#use strict;

use warnings;
use Bio::DB::GenBank;
use Bio::SeqIO;
use Bio::SeqFeatureI;

#open my $OUTFILE, '>', "Gb_parser.fasta";
my $gb = new Bio::DB::GenBank; 
# Récupération dans $seq de l'objet Genbank contenant de nombreuses informations
# Gi, Acc, séquence, Annotations, Organisme, Espèce, Genre ...

my @Acc = qw(ABK27677); #Accession number de la séquence à traiter
my $seq1 = $gb->get_Seq_by_acc(@Acc);
my $Sequence = $seq1->seq();
my $Description = $seq1->desc();
print "Acc = @Acc\nDescription = $Description\n"; #ecrit le num d'accession et sa description associée
my $seqio_object = $seq1;

for my $feat_object ($seqio_object->get_SeqFeatures) {
    #print "\n", $feat_object->primary_tag, "\n";
    for my $tag ($feat_object->get_all_tags){
        print $tag." ";
        for my $value ($feat_object->get_tag_values($tag)){
            print  $value."\n"      
        }
    }
}

foreach $feat ( $seqio_object->get_SeqFeatures() ) {
if ($feat->primary_tag() eq "CDS"){ # ne s'intéresse qu'aux tags "CDS"
    foreach $tag ( $feat->get_all_tags() ) {    
        if ($tag eq "locus_tag"){
            print  "\n>the locus: ", join(' ',$feat->get_tag_values($tag)); #ecrit le locus
        }
        elsif ($tag eq "product"){
            print  " ", join(' ',$feat->get_tag_values($tag)); #ecrit le nom du CDS
        }
        elsif ($tag eq "coded_by"){
            print "\n\nThe corresponding gene acc: ", $feat->get_tag_values($tag);
            #print ">\n\n", join (' ', $feat-> get_tag_values($tag));     #affiche l'acc. du gene

        }
    }

     #print "\n"
    print " \n\n", $feat->start, "..", $feat->end,"\n\n";;# " on strand ",     $feat->strand, "\n";
    $out = $feat->seq; #out est un Primary::Seq
    $string = $out->seq(); #recupère la séquence
    print $string."\n\n";
}
}

There is 2 problem with my script:

1. I can't use it with several accession numbers (IDs)
2. The goted CDS sequences are protein sequences, in my case I'm looking for the corresponding nucleic sequences. We have just the coded_by infos that I don't know how I can use it to get the nucleic sequence