Hi Abhishek

If you want every coding transcript for each gene you can create an if loop to select only protein coding transcripts. Also, you can use the coding_region_start and coding_region_end calls on the transcript to get these.

while (my $gene_id = shift @gene_ids) {

my $gene = $gene_adaptor->fetch_by_stable_id($gene_id);

my @transcripts = @{$gene->get_all_Transcripts};

while (my $transcript = shift @transcripts){

    if ($transcript->biotype eq 'protein_coding' {
        my $cdsseq = $transcript->translateable_seq;
        my $cdsstart = $transcript->coding_region_start;
        my $cdsend = $transcript->coding_region_end
        my $protseq = $transcript->translate->seq;
        print #stuff
        }
    }
}

See http://www.ensembl.org/info/docs/Doxygen/core-api/classBio_1_1EnsEMBL_1_1Transcript.html#a29a5039bad42ce6939b42cfeef1e3b4e

So, I know this is old, but haven't found another thread more related to my problem... I have being trying to obtain a fasta file of either cDNA or CDS sequences of only protein coding genes and only of the canonical transcript. For that, I was checking if the translatable sequence was defined for a given gene, and then print the corresponding sequence. But it doesn't matter how or where I set the if (defined ....), that the result is always a list with a canonical transcript for ALL genes of a given species. I was trying to use my script to retrieve the sequences for example of Nematostella vectensis. The command I run was:

perl script.pl nematostella_vectensis 1 cdna

Where with 1, I am setting 1 bp as the min length to retrieve a seq (i.e., all of protein coding ones) and with "cdna", the whole transcript.

And the script's code:

#!/usr/bin/perl -w

use strict;
use warnings;

use Bio::EnsEMBL::Registry;
use Data::Dumper;

my $species=$ARGV[0];
my $minlength=$ARGV[1];
my $seqtype=$ARGV[2];
if ($seqtype ne "cds" && $seqtype ne "cdna"){
    print "Select the type of sequence you want to retrieve: cdna or cds. \"$seqtype\" is not either.\nUsage: perl retrieve_protcod_canonicaltranscript speciesname mintranscriptlength cdna/cds\nExample: perl retrieve_all_canonicaltranscript pelodiscus_sinensis 200 cdna\n\n";
    exit;
}

my $registry = "Bio::EnsEMBL::Registry";
$registry->load_registry_from_multiple_dbs(
    {-host => 'mysql-eg-publicsql.ebi.ac.uk',
     -port => 4157, 
     -user => 'anonymous'
    },
    {-host => 'ensembldb.ensembl.org',
     -port => 5306,
     -user    => 'anonymous'
    }
);

unless (open(OUTPUT, ">${species}_protcod_canonicaltranscript$seqtype.fa")){
    print "Cannot create an output file\n\n";
    exit;
}

my $gene_adaptor  = $registry->get_adaptor( $species, 'Core', 'Gene' );
my $transcript_adaptor = $registry->get_adaptor( $species, 'Core', 'Transcript');
my @gene_ids= @{$gene_adaptor->list_stable_ids()};

while (my $gene_id = shift @gene_ids) {
    my $gene = $gene_adaptor->fetch_by_stable_id($gene_id);
    if (defined ( my $cds = $gene->canonical_transcript()->translateable_seq())){
        my $cdna = $gene->canonical_transcript()->spliced_seq();
        if (length $cdna >= $minlength && $seqtype eq 'cdna') {
            print OUTPUT ">$gene_id\n$cdna\n";
        } elsif (length $cds >= $minlength && $seqtype eq 'cds') {
            print OUTPUT ">$gene_id\n$cds\n";
        } else {
            next;
        }
    }
}
exit;