Exons:
Relevant bits of information for using the MUSC Shared Resource,
 the BioMolecular Computing Resource (BCR).

Exon_55 Hidden Markov Model v 2.1.1 at MUSC.

This software written by Sean Eddy at USTL is really a suite of 9 programs 
which produce calibrate and utilize Hidden Markov Model algorithms for 
multiple sequence analysis.

The programs are:

hmmalign- aligns a sequence to an HMM
hmmbuild- builds an HMM from an alignment
hmmcalibrate- calibrates the HMM produced by hmmbuild
hmmconvert-
hmmemit-
hmmfetch
hmmindex-
hmmpfam-
hmmsearch-

This software accepts as input multiple sequence file (MSF) alignment 
(pileup or CLUSTALW formats work) and contructs a position specific 
matrix based on Hidden Markov Model procedures which describes the alignment. 
The output is an HMM which can be used to scan additional sequences 
for their similarity to the parent set.

The programs may be executed by typing the program name. Typing the option
"-h" will present a brief usage summary.

hmmbuild -h  

Would cause the program to echo a usage summary for hmmbuild. 

A typical HMMER series would be to build the HMM from an MSF with hmmbuild. 
Next one would calibrate the HMM using hmmcalibrate. Next hmmsearch might be 
used to align a new sequence to the HMM. hmmsearch can also use the
HMM to search through a database. This can be GCG formatted database
of your own or it may be one of the BCR archives. When you are ready to
try database searching with HMMER send an email to 
Starr Hazard for the details
about the proper path to direct hmmsearch to use.  

     
 

The BCR at the Medical University of South Carolina
URL: http://bcr.musc.edu/
Last modified: MArch 12, 2001- ESH

e-mail to Starr comments about this page.