[ensembl-dev] Reconstructing mRNA from transcript coordinates (human)

[Allan Kamau]

I would like to search for a predetermined non-palindromic motif at
the 3' end of immature (pre) mRNA for specific human genes. Some of
these genes are found on the negative strand.
So far I have managed to obtain the nucleotide sequences using the
coordinates provided by the transcript object as shown at the end of
these text.

How do I now "convert" these sequences into nucleotide representations
of mRNA. Where the left most nucleotide will be at the 5' end and the
right most nucleotide will be at the 3' end.

1)For positive strand genes - should I take reverse complement of
these sequences.

2)For negative strand genes - should I use these sequences as they are.



Below is the code to obtain the transcript coordinates and the
nucleotide sequence using these readings.

    my $stable_id  = $feature->stable_id();
    my $seq_region = $feature->slice->seq_region_name();
    my $start      = $feature->start();
    my $end        = $feature->end();
    my $strand     = $feature->strand();
    my $slice=$slice_adaptor->fetch_by_region('chromosome',$seq_region,$start,$end,$strand);
	my $seq="";
    if($slice)
    {
		$seq=$slice->seq();
	
		print sprintf("%s: %s:%d-%d (%+d)",
			$stable_id, $seq_region, $start, $end, $strand
		);
}