(97) Successive State Splitting of Protein Hidden Markov Network

	Machine:     UNIX machine
	Environment: UNIX, Motif
	Language:    C
	Source Code: 0.1 MB
	Documents:   Manual (English)

Overview

Successive State Splitting is an optimization mechanism of hidden Markov networks using the same expectation maximization criteria as the Hidden Markov Model (HMM) learning.

Features

the HMM engine, which controls HMM (Viterbi) learning,
the network generator, which generates a new network by splitting the chosen state,
the network tester, to evaluate the efficiency of the splitting and to determine which state to split and when to stop the algorithm, and
the GUI Monitor, which shows the network shape as well as every output and transition probabilities.

Function

This system provides a hidden Markov model for a set of amino acid sequences of proteins.
This system provides an optimal network and hidden Markov model at the same time using the successive state splitting algorithm.
Network optimization and HMM learning are conducted under the same criteria: expectation maximization.

FTP

README,
DOC.
Successive State Splitting of Protein Hidden Markov Network [24K]

www-admin@icot.or.jp