Bayesian neural networks for prediction of protein secondary structure

Authors:
Jianlin Shao;Dong Xu;Lanzhou Wang;Yifei Wang
Affiliations:
College of Life Sciences, China Jiliang University, Hangzhou, Zhejiang Province, China;College of Sciences, Shanghai University, Shanghai, China;College of Life Sciences, China Jiliang University, Hangzhou, Zhejiang Province, China;College of Sciences, Shanghai University, Shanghai, China
Venue:
ADMA'05 Proceedings of the First international conference on Advanced Data Mining and Applications
Year:
2005

Citing 4
Cited 0

A practical Bayesian framework for backpropagation networks

Neural Computation
Suppressing random walks in Markov chain Monte Carlo using ordered overrelaxation

Learning in graphical models
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks
Convergence assessment techniques for Markov chain Monte Carlo

Statistics and Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

A novel approach is developed for Protein Secondary Structure Prediction based on Bayesian Neural Networks (BNN). BNN usually outperforms the traditional Back-Propagation Neural Networks (BPNN) due to its excellent ability to control the complexity of the model. Results indicates that BNN has an average overall three-state accuracy Q3 increase 3.65% and 4.01% on the 4-fold cross-validation data sets and TEST data set respectively, comparing with the traditional BPNN. Meanwhile, a so-calledcross-validation choice of starting values is presented, which will shorten the burn-in phase during the MCMC (Markov Chain Monte Carlo) simulation substantially.