In article <6l3pas$sbk at net.bio.net>,
David R. Bell <david.bell at nottingham.ac.uk> wrote:
>I am trying to use Molphy- protml- for ML analysis of proteins, but I wish to
>use Blosum62 instead of JTT.
>Does anyone know how to format the matrix and use it to score the protml
>program ?
JTT is a probability model of protein evolution whose probabilities are
estimated empirically, as PAM is too.
When one has such a model, it is possible to approximately derive
alignment weights from it by taking logs of odds and rounding them off.
BLOSUM is such a rounded-off, logged, set of weights. Unfortunately however
great it may be, it did not start as a probability model of protein evolution.
So it can't be used in any protein maximum likelihood program. So this is
not just a problem in file formatting, it goes deeper than that.
If I am wrong about this, I hope someone will point out that there is a
BLOSUM62 model (rather than a set of weights).
--
Joe Felsenstein joe at genetics.washington.edu
Dept. of Genetics, Univ. of Washington, Box 357360, Seattle, WA 98195-7360 USA
