Manisha Goel asked > I was trying to develop an algorithm for describing/predicting a > pattern (e.g. transmembrane region, signal peptide etc) in protein > sequences. I want to derive this pattern from the multiple sequence > alignments. But I was wondering if I should use substitution matrices > or HMMs to describe/represent these patterns. Are there any definite > advantages of using one over the other? Does the choice depend on what > I am trying to define? Can somebody please direct me to relevant > literature or suggest something from personal experience? HMMs are currently the best method for representing patterns of the type you have in mind. Profile HMMs are the most popular, and are supported by two main packages HMMer and SAM. Both packages are free to academics, non-profits, and government researchers, but the HMMer package is open-source and SAM is not (at least not yet---we're thinking of making it open-source but have not had the time or resources to clean up the source code enough to do that reasonably). SAM and HMMer models are slightly different, but similar enough to be interconvertible with only fairly small losses (SAM models are slightly more general than HMMer models, and use a different calibration method). There is sam2hmmer and hmmer2sam software available on the web. For developing new profile HMMs, SAM is a better choice, because there has been more development on the model-building code. (HMMer is more popular than SAM, largely because of the prebuilt PFAM resource, which is a very valuable database.) See http://stash.mrc-lmb.cam.ac.uk/HMMER-SAM/ for information about a test comparing HMMER and SAM, be people who were not on the development team for either and were trying to decide which to use. If you want to do non-profile HMMs (such as the transmembrane models of TMHMM), then you may have to build your own code---I've not seen general-purpose HMM code that was a good utility kit for building new HMMs. Of course, I haven't been looking for one, so I may have missed some major developments. ------------------------------------------------------------ Kevin Karplus karplus at soe.ucsc.edu http://www.soe.ucsc.edu/~karplus Professor of Biomolecular Engineering, University of California, Santa Cruz Undergraduate and Graduate Director, Bioinformatics Senior member, IEEE Board of Directors, ISCB (starting Jan 2005) life member (LAB, Adventure Cycling, American Youth Hostels) Effective Cycling Instructor #218-ck (lapsed) Affiliations for identification only.