Yes, they're all over the place and lend themselves well to nucleotide and protein sequence data. Hidden Markov Models are used in the Glimmer gene finding algorithm for example.