Bayesian Testing for Multiple Motif in Biological Sequences Based on Moment Estimate
For the motif significant testing in biological sequences, Bayesian testing based on moment estimate is presented.The motif significant testing is converted into the goodness of fit test of the multinomial distribution.While the prior distribution of the multinomial distribution is known as Dirichlet, the estimates of hyper parameters of prior distribution are given using moment estimate.Based on Bayesian Theorem, a Bayes factor is obtained, which acts as statistical estimation of the significance of test.The method overcomes the difficulty of constructing the test statistic and deriving its exact distribution on the null hypothesis.Taking correlation coefficient as an objective criterion of the quality, experimental results indicate that our Bayesian testing performed better on average than the classical methods, such as the shifted fast Fourier transform (sFFT) and the cyclic shifted fast Fourier transform ( csFFT).
bioinformatics multinomial distribution Bayesian Theorem prior distribution hyper parameter moment estimate
Qian Liu Sanyang Liu Lifang Liu
School of Science Xidian University Xian,China School of Computer Science and Technology Xidian University Xian,China
国际会议
北京
英文
1-4
2009-06-11（万方平台首次上网日期，不代表论文的发表时间）