GCG: distances changing gap weights

David Mathog mathog at seqvax.caltech.edu
Sat Aug 27 12:30:00 EST 1994


In article <Cv5rJ5.Hqt at news.cis.umn.edu>, nrd at insitu.med.umn.edu (Neal Dalton) writes...
>We would like to change the weight given to gaps vs nucleotides when
>calculating distances using GCG's distance program.  Currently, gap

Why, and to what?

>mismatches are given the same weight as nucleotide mismatches.

Well, yes for A vs. G, but no for T vs. Y, for instance.  Note that that
weight is zero, so there need not be an explicit "." column in the table
as GCG specifies that if the symbol isn't in the table, any comparison
against it returns zero. 

> 
>How can we do this?  I tried to add gaps to the dnadistances.cmp table,
>but that didn't seem to work.
> 

It should have worked, take a look at the randomdna.cmp file for example.
Possibly you need to run it through reformat to clean it up?

On the off chance that there is some bug that precludes using a "." column/
row you could edit the .MSF file that you're working on and convert all
gaps to some other symbol, for instance, "Y".  Then edit a *copy* of the
comparison table, changing all of the values in the Y associated columns/
rows to be what you want for A vs. gap, G vs. gap, etc.  This assumes that
you don't have any "Y" entries in your data. 

Regards,

David Mathog
mathog at seqvax.bio.caltech.edu
Manager, sequence analysis facility, biology division, Caltech 




More information about the Bio-soft mailing list