In article <Pine.PMDF.3.91.951110082911.539480474A-100000 at crcvms.unl.edu>,
bjadams at CRCVMS.UNL.EDU wrote:
> Concerning the Determination of Homologous Sequences,
> David Maddison wrote:
there's
> <just nothing to get a hold of."
>>Try the concept of looking at grouped content.
More similar sequence will by nature of both DNA and peptide sequence
have simlar frequencies of n-plets.
Clustering according to this method can produce remarkably accurate
distance measures, but if the data is simply divergent protein noise,
then a more rigorous di-peptide freequency analysis could provide
some help. Basically though noise in noise out.
Win
--
All views are via telescope...
