Interpolated hidden markov models estimated using conditional ML for eukaryotic gene annotation

Authors:
Hongmei Zhu;Jiaxin Wang;Zehong Yang;Yixu Song
Affiliations:
State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing, China;State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing, China;State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing, China;State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing, China
Venue:
ICIC'06 Proceedings of the 2006 international conference on Computational Intelligence and Bioinformatics - Volume Part III
Year:
2006

Citing 1
Cited 0

Two Methods for Improving Performance of a HMM and their Application for Gene Finding

Proceedings of the 5th International Conference on Intelligent Systems for Molecular Biology

Quantified Score

Hi-index	0.00

Visualization

Abstract

To improve the performance of computational gene annotation programs, we introduced the well known interpolated Markov Chain (IMC) technology to the class Hidden Markov models (CHMM). CHMM was applied in one of the best eukaryotic gene prediction systems: HMMgene. The conditional Maximum Likelihood estimation (CMLE) algorithm was educed to estimate the interpolation parameters. The resulting gene prediction program improves exon level sensitivity by 3% and specificity by about 1% compared to HMMgene as trained and tested on some standard human DNA sequence dataset.