[EMBOSS] Tranalign relaxation?

[Justin Havird]

Thanks,

Yes the problem is that some taxa don't follow the standard genetic codes
available for some codons, usually start codons. I may try to pass the
sequences to tranalign without the first codon and see if that will clear up
a majority of the troublemakers. I thin the issue with "unambigouous
ambiguous codons" is what is happening in cases where there is an ambiguous
nucleotide position, even though the amino acid should be able to be
determined.

Thanks again and let me know if anything else comes to mind,

Justin

On Mon, May 31, 2010 at 6:08 PM, Peter <biopython at maubp.freeserve.co.uk>wrote:

> On Mon, May 31, 2010 at 11:38 PM, Justin Havird <jchavird at gmail.com>
> wrote:
> > Hi Peter,
> >
> > Yes, I have been using the correct mitochondrial translation codes.
> > Depending on the taxonomic group, there are about 4 different
> mitochondrial
> > translation codes that I have been using. The problem is that some
> > individual species use alternate start codons, etc. These are generally
> > case-by-case problems.
> >
> > Someone suggested I alter the tranalign program itself and create a new
> > translation table that would take these alternate start codons into
> > consideration. However, I haven't had much luck with this.
> >
> > Any other ideas?
> >
> > Thanks!
> >
> > Justin
>
> If I understand correctly, part of the problem is your organisms
> don't all fully follow the standard genetic codes available in
> EMBOSS. Therefore would a new tranalign option to give the
> start codon(s) to use be helpful?
>
> Personally I'd probably solve this kind of thing by writing a script
> in (Bio)python. I might still call tranalign to do the hard work but
> might try passing it the sequences with the first codon/amino
> acid removed.
>
> You didn't CC the mailing list - perhaps an accident? Please
> feel free to forward my reply back to the list.
>
> Peter C.
>
> P.S. There were some issues with using EMBOSS transeq for
> the translation of "unambigouous ambiguous codons"  which I
> reported back in July 2009:
> http://lists.open-bio.org/pipermail/emboss/2009-July/003667.html
>
> I haven't been back to test if the latest EMBOSS release
> addressed it (EMBOSS 6.1.0 is still broken in this regard
> I think), but your problem with ambiguous IUPAC codes
> may be linked to this.
>