Mamot

MAMOT (hidden MArkov MOdelling Tool) is a program that implements the classic algorithms used for HMM modelling with some additional non-standard options.

The user specifies the model and the parameters in convenient input files. File output is generated when appropriate.

Available are the following functionalities:
1) generation of random sequences according to the specified model,
2) Baum-Welch (BW, EM) LEARNING,
3) Viterbi LEARNING,
4) FB (forward(-backward)) PROBABILITY of a sequence given the model,
5) FB forward-backward "posterior" DECODING, and
6) Viterbi probability and DECODING.

At this moment we make available executables:
compiled to run under LINUX, version of September 2006:
mamotlinux084
compiled to run under LINUX, version of June 2006:
mamotlinux092

Useful information is given in:

README084
resp.
README092
files.

Enquiries can be addressed to:
mauro.delorenzi@isrec.unil.ch
and to
Frederic.Schutz@isb-sib.ch

For an example application to transcription factor binding site recognition, see
Example