论文信息 - Extraction of hidden Markov model representations of signal patterns in DNA sequences.

Extraction of hidden Markov model representations of signal patterns in DNA sequences.

We have developed a method to extract the signal patterns in DNA sequences. In this method, the Genetic Algorithm (GA) and Baum-Welch algorithm are used to obtain the best Hidden Markov Model (HMM) representations of the signal patterns in DNA sequences. The GA is used to search the best network shapes and the initial parameters of the HMMs. Baum-Welch algorithm is used to optimize the HMM parameters for the given network shapes. Akaike Information Criterion (AIC), which gives a criterion for the balance of adaptation and complexity of a model, is applied in the HMM evaluation. We have applied the method to the extraction of the signal patterns in human promoters and 5' ends of yeast introns. As a result, we obtained HMM representations of characteristic features in these sequences. To validate the efficiency of the method, we have performed promoter recognition using obtained HMMs. Two entries including nine promoters are selected from GenBank 76.0, and it is observed that the HMM can predicts eight promoters correctly. These results imply that the method is efficient to design preferable HMM networks, and provides reliable models for the recognition of the signal patterns.

[1] L. R. Rabiner,et al. An introduction to the application of the theory of probabilistic functions of a Markov process to automatic speech recognition , 1983, The Bell System Technical Journal.

[2] D. Haussler,et al. Protein modeling using hidden Markov models: analysis of globins , 1993, [1993] Proceedings of the Twenty-sixth Hawaii International Conference on System Sciences.

[3] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[4] Andrzej K. Konopka. PLAUSIBLE CLASSIFICATION CODES AND LOCAL COMPOSITIONAL COMPLEXITY OF NUCLEOTIDE SEQUENCES , 1993 .

[5] H. Akaike,et al. Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[6] A. Stewart. Genes V , 1994 .

[7] A. Konagaya,et al. Stochastic motif extraction using a genetic algorithm with the MDL principle , 1993, [1993] Proceedings of the Twenty-sixth Hawaii International Conference on System Sciences.

[8] Akihiko Konagaya,et al. Stochastic Motif Extraction Using Hidden Markov Model , 1994, ISMB.

[9] Hidetoshi Tanaka,et al. Classification of Proteins via Successive State Splitting Algorithm of Hidden Markov Network , 1993 .

[10] P. Bucher. Weight matrix descriptions of four eukaryotic RNA polymerase II promoter elements derived from 502 unrelated promoter sequences. , 1990, Journal of molecular biology.