Abstract— Protein structure prediction is a challenging field strongly associated with protein function and evolution determination, which is crucial for biologists and the pharmaceutical industry. Despite significant process made in recent years, protein structure prediction maintains its status as one of the prime unsolved problems in computational biology. In this study, we have developed a method for protein structure prediction based on profile Hidden Markov Model (HMM) and Quantum Particle Swarm Optimization (QPSO) with diversity-maintained algorithm (DMQPSO). The profile HMM can reduce the number of states using secondary structure information about proteins for each fold, which is called a 7-state HMM. The DMQPSO is an efficient optimization algorithm which is used to train profile HMM. Experiment results show that the proposed method is reasonable and the accuracy of protein secondary structure prediction is increased.
Index Terms— Protein structure prediction, protein secondary structure, fold recognition, profile HMM, DMQPSO.
The authors are with School of Information Science Technology, Hainan Normal University, Haikou 571158, HaiNan, China (e-mail: haixia_long@163.com, 595615374 @qq.com, 605515770 @qq.com).
[PDF]
Cite: Haixia Long, Shulei Wu, and Chun Shi, " Protein Structure Prediction Based on Profile HMM and DMQPSO," International Journal of Information and Electronics Engineering vol. 5, no. 4, pp. 280-285, 2015.
