Extremely high substitution rates

[Vazrick Nazari]

Hello all,
An interesting problem that I thought might find a
solution if shared here.

I used Modeltest 3.06 in conjunction with PAUP* 4.0 to
estimate the best substitution model for sequence data
of 7 genes I am using for a phylogenetic analysis. For
each dataset, Modeltest predicts the best fit model,
base frequencies and substitution rates, as well as
gamma distribution shape parameter and proportion of
invariable sites.

It worked fine in every case except COII, where it
predicted extremely high substitution rates (e.g. A-C:
6.9646e+06, etc.). I repeated the test several times
using different versions of PAUP*, Modeltest, and
deleted a short region of indels in the dataset to see
if it will work right - it didn't.

David Posada, the author of Modeltest, thinks this is
a bug in PAUP*. i tried contacting PAUP people but
haven't got any response from them yet.

Has anybody ever experienced such a problem? Is there
a solution?
I can provide more information for those who might
have an answer to this.

Cheers,

Vazrick Nazari
Graduate Student
Department of Biological Sciences
CW 405, Biological Sciences Centre
University of Alberta
Edmonton, Alberta
T6G 2E9  Canada



__________________________________________
Yahoo! DSL – Something to write home about.
Just $16.99/mo. or less.
dsl.yahoo.com