摘要
We propose a novel model to predict RNA secondary structure based on the fuzzy sets theory. Through the fuzzy partition of state spaces and the incorporation of fuzzy goals, we can find the optimal fuzzy policy of the model using fuzzy dynamic programming algorithm effectively, and then determine optimal and suboptimal RNA secondary structures. Compared to the existing sophisticated prediction models, such as Zuker＇s method and the SCFG model, our fuzzy model based approach has many advantages： 1） computational complexity can be reduced by the fuzzy partition; 2） the optimal secondary structure and several suboptimal ones can be generated simultaneously; and 3） subjective prior knowledge can readily be incorporated. This paper presents a complete description of our fuzzy model and gives the implementation of the proposed method. We also apply the BJK fuzzy model structure to secondary structure predictions based on datasets of tRNA and tmRNA sequences. By the comparison of our fuzzy method with both the minimum free energy based mfold tool and the BJK grammar model of SCFG, our experimental results validate the effectiveness of the proposed method and the prediction accuracy is shown to be further improved.
We propose a novel model to predict RNA secondary structure based on the fuzzy sets theory. Through the fuzzy partition of state spaces and the incorporation of fuzzy goals, we can find the optimal fuzzy policy of the model using fuzzy dynamic programming algorithm effectively, and then determine optimal and suboptimal RNA secondary structures. Compared to the existing sophisticated prediction models, such as Zuker＇s method and the SCFG model, our fuzzy model based approach has many advantages： 1） computational complexity can be reduced by the fuzzy partition; 2） the optimal secondary structure and several suboptimal ones can be generated simultaneously; and 3） subjective prior knowledge can readily be incorporated. This paper presents a complete description of our fuzzy model and gives the implementation of the proposed method. We also apply the BJK fuzzy model structure to secondary structure predictions based on datasets of tRNA and tmRNA sequences. By the comparison of our fuzzy method with both the minimum free energy based mfold tool and the BJK grammar model of SCFG, our experimental results validate the effectiveness of the proposed method and the prediction accuracy is shown to be further improved.
基金
the National Natural Science Foundation of China (Grant No. 60621062)
Teaching and Research Award Program for Out-standing Young Teachers in Higher Education Institutions of MOE (TRAPOYT), China
