arlin at is.dal.ca
Tue Feb 20 21:59:41 EST 1996
I'm imagining a tool called "homologizer" that would be used
to map features of homologs onto a target sequences using
an alignment. I would feed the homologizer a bunch of
homologous sequences A, B, C etc. and any associated GenBank
features tables, Swiss-Prot annotations, PDB files, etc.
This might include a new sequence X whose function is of interest.
After the homologizer calls an alignment program to align the
sequences, I can click on X (or some other sequence), and the
homologizer would present me with a clickable map (for instance)
in which *all* of the features were mapped onto sequence X.
Clicking on a feature would take me back to a piece of
information (from a GenBank, Swiss-Prot, PDB, etc entry)
with more details.
The homologizer would be useful to many researchers using
molecular sequences, since one of the most common inferences
that we make is to align a new sequence with previously
characterized ones, and then make inferences about the
functions of sites and regions in the new sequence
based on the properties associated with other sequences
in the alignment. Such a tool could be used to plan
site-directed mutagenesis or to refine an alignment based
on structural principles or to analyze the evolution of
a set of related proteins. For some work that I am doing,
I would like to map intron positions from homologous
protein-coding genes onto a single gene.
A homologizer could also be incorporated
into database schemes to allow queries like "are there
any sites in my input sequence that align with active
site residues in a homolog?" (and much more).
Probably the 'homologizer' doesn't exist, but does anyone
know of a tool (maybe a specialized alignment browser) that
does at least *some* of this?
Arlin Stoltzfus (arlin at is.dal.ca)
Department of Biochemistry
Halifax, NS B3H 4H7 CANADA
ph. 902-494-3569; fax 902-494-3569
More information about the Bio-soft