Approaches to gene finding: Hidden Markov models

• Model

  • A finite model describing a probability distribution over all possible sequences of equal length
  • “Natural” scoring function
  • (Conditional) Maximum likelihood “training”

• Markov

  • k-order Markov chain: current state dependent on k previous states
  • The next state in a 1st-order Markov model depends on current state

• Hidden

  • Hidden states generate visible symbols

• Assumptions

  • Independence of states
    • No long range correlation

• Example: HMMgene, A. Krogh (1998), In Guide to Human Genome Computing, 261-274.

Previous slide

Next slide

Back to first slide

View graphic version