In article <8hgvoq$obc at gap.cco.caltech.edu>, mathog at seqaxp.bio.caltech.edu (David Mathog) writes:
|> 2. There's some sort of memory problem feeding hmmsearch a really
|> large translated database in this manner - I only just today wrote to S. Eddy
|> about that so don't know the solution yet. It may or may not affect
|> Unix systems. I suspect it will cause the hmmsearch process to use a
|> lot of memory though. I also suspect that the problem is related to
|> the huge "protein" fasta entries which result from just translating
|> an entire genbank DNA entry. Since it seems to depend more on the
|> composition of the data fed in than on the absolute size of the
|> database.
The SAM package provides many of the same features as the HMMer
package (and a number of features not available with HMM). We have
had no problem with SAM running large databases through the search
programs, probably because SAM uses a space-efficient version of the
fundamental dynamic programming algorithms.
SAM is free to academics and non-profits and is available at
reasonable license fees to corporations.
See
http://www.cse.ucsc.edu/research/compbio/sam.html
for more information.
You can also try out some of the uses of SAM for free at
http://www.cse.ucsc.edu/research/compbio/HMM-apps/HMM-applications.html
--
Kevin Karplus karplus at cse.ucsc.eduhttp://www.cse.ucsc.edu/~karplus
life member (LAB, Adventure Cycling, American Youth Hostels)
Effective Cycling Instructor #218-ck
Anything below this line is junk added by others without my approval.
