| Paper: | MLSP-P2.2 | ||
| Session: | Bioinformatics and Biomedical Applications | ||
| Time: | Wednesday, May 19, 13:00 - 15:00 | ||
| Presentation: | Poster | ||
| Topic: | Machine Learning for Signal Processing: Bioinformatics Applications | ||
| Title: | PROTEIN SECONDARY STRUCTURE PREDICTION WITH SEMI MARKOV HMMS | ||
| Authors: | Zafer Aydin; Georgia Institute of Technology | ||
| Yücel Altunbaşak; Georgia Institute of Technology | |||
| Mark Borodovsky; Georgia Institute of Technology | |||
| Abstract: | Secondary structure prediction has been an essential task indetermining the structure and function of the proteins. Prediction accuracy is improving every year towards the 88\% estimated theoretical limit~\cite{Rost2002}. There are two approaches for the secondary structure prediction. The first one, {\it ab initio} (single sequence) prediction does not use any homology information. The evolutionary information, if available, is used by the second approach to improve the prediction accuracy by a few percentages~\cite{FRISHMAN1997}. In this paper, we address the problem of single sequence prediction by developing a semi Markov HMM, similar to the one proposed by Schmidler {\it et al.}~\cite{SCHM2000}. We introduce a better dependency model by considering the statistically significant amino acid correlation patterns at segment borders. Also, we propose an internal dependency model considering “right to left” dependencies without modifying the “left to right” HMM topology. In addition, we propose an iterative training method to better estimate the HMM parameters. Putting all these together, we obtained 1.5\%improvement in three-state-per-residue accuracy. | ||
| Back | |||
Home -||-
Organizing Committee -||-
Technical Committee -||-
Technical Program -||-
Plenaries
Paper Submission -||-
Special Sessions -||-
ITT -||-
Paper Review -||-
Exhibits -||-
Tutorials
Information -||-
Registration -||-
Travel Insurance -||-
Housing -||-
Workshops