A Balanced Comparative RNA Secondary Structure Prediction Method Using Weighted Adaboost Extreme Learning Machine
RNA secondary structure prediction is an open problem in bioinformatics.Comparative sequence analysis method is widely used in RNA secondary structure prediction.It takes a group of related RNA molecules as input, and predicts the consensus structure, which is able to increase the prediction accuracy and overcome the problem of pseudoknots.However, When using this kind of method, users always encounter the problem of data imbalance, which means there are a few positive samples (base pairs), but much more negative samples in the data set.This will lead to losing some meaningful potential base pairs.To solve the problem, we propose a method called weighted Adaboost ELM to solve the data imbalance,which merge the weighted Adaboost framework and extreme learning machine based on comparative sequence analysis method.The proposed method is compared with extreme learning machine, supporter vector machine and RNAalifold, in metrics of sensitivity, specificity, Matthews correlation coefficient (MCC) and g-mean, based on 68 RNA aligned families from Rfam, version 11.0.The results show that the proposed method performs best in sensitivity and g-mean, which means it is good at disposing imbalanced data, while does not have superiority in specificity and MCC.
RNA secondary structure prediction Comparative sequence analysis Covariation score Weighted Adaboost Extreme learning machine Imbalanced data
Tianhang Liu Jianping Yin
School of Computer Science,National University of Defense Technology,Changsha 410073,China
国内会议
金华
英文
1-9
2015-10-30（万方平台首次上网日期，不代表论文的发表时间）