基于改进适应度函数的RNA结构预测

RNA二级结构预测是生物信息学的重要研究内容。本文提出了一个新的启发式算法进行带假节的RNA结构预测。本文首先通过对RNA序列的若干特征和RNA二级结构进行相关性分析，从中选择跟RNA结构有较大相关性的特征，然后依据遗传算法、综合自由能、被选择茎区的条数以及被选择茎区的平均长度等特征来构造打分函数预测RNA的结构。本文对该方法进行了测试，结果表明本文所采用的从特征分析中得出的打分函数以及通过启发式算法来叠加茎的方法是有效的,在对tRNA以及5SrRNA等序列的预测上相比单纯的自由能最小方法有更高的准确性。并且该方法进一步推广到预测含假结的RNA的二级结构时也有较好的结果。

RNA structure prediction based on improve the fitness function

RNA secondary structure prediction is an important research field of bioinformatics. In this paper we propose a new heuristic method to predict RNA secondary structure including pseudoknot. Based on the analysis of specific features of RNA structures, we select the features that have significant relationship with RNA structure characteristics. Then we use the genetic algorithm and incorporate those features: the free energy, the number and the length of selected stems to predict the RNA structure. The experiment results show that the scoring function concluded from the feature analysis as well as the heuristic stem adding algorithm is effective. It outperforms the traditional minimal free energy algorithm on tRNA and 5SrRNA data. This method also has good prediction results on the pseudo-knot structure.