JalView and Phylip output

Hi Sebastian.

Thanks for the email. I've forwarded this to the jalview-discuss list
because I suspect this is of general interest.

Sebastien Moretti wrote:

Do you plan to add a phylip format output to Jalview ?

it wasn't planned originally. Our preferred route for phylogeny
calculations was to pass the alignment to TOPALi (www.topali.org) or use
phylogeny web services (which don't exist as yet).

I know that there is a 'problem' with the fact that strict phylip format
must have sequence names cut to the 10th characters.

yep - its only a problem if you care about the names - there would have
to be some user investigation concerning that. jalview doesn't need
unique names (in principal), but the tree-matching display requires
unambiguous correspondence. If phylip analyses were implemented via a
web service then all of that kind of housekeeping could be hidden away
from the user.

But, now, more and more software, like PhyML or PAML, use names longer
than 10: 30 for PAML, no limit (?) for PhyML.

true. They still require unique names, however.

What do you think to add a strict phylip format output and a non-strict
one ?

It would be of use - but a switch would be needed between the strict and
non-strict when generating a PHYLIP formatted aligment. Otherwise, from
the looks of the format,
(http://evolution.genetics.washington.edu/phylip/doc/sequence.html) it
would be a very easy implementation since it is practically the same as
PFAM (except for the interleaved/not interleaved aspect).

Do you know when the release with AMSA format support will be available ?

The jalview paper is still being written, and we originally planned to
delay the release of 2.4 until after it was submitted. However, I
suspect the 'official' release may happen before that now.

thanks for the mail!


J. B. Procter (ENFIN/VAMSAS) Barton Bioinformatics Research Group
Phone/Fax:+44(0)1382 388734/345764 http://www.compbio.dundee.ac.uk
The University of Dundee is a Scottish Registered Charity, No. SC015096.