Structural alignment of long, highly structured RNAs based on ordered tree method
Konferenz: BIBE 2019 - The Third International Conference on Biological Information and Biomedical Engineering
20.06.2019 - 22.06.2019 in Hangzhou, China
Tagungsband: BIBE 2019
Seiten: 5Sprache: EnglischTyp: PDFPersönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Gao, Longjian; Xu, Chengzhen; Sun, Xueqin; Song, Wangan; Xiao, Feng (School of Computer Science and Technology, Huaibei Normal University, Huaibei, Anhui, China)
Zhang, Dongbing (School of Computer Science and Technology, Huaibei Normal University, Huaibei, Anhui, China & Information College, Huaibei Normal University, Huaibei, China)
Based on the characteristics of long non-coding RNA (lncRNA) structure with long sequence and high complexity, we here propose a new lncRNA structure alignment algorithm based on the combination of optimization algorithm and neural network algorithm. This algorithm can transform complex RNA structure into simple ordered tree, which is more convenient for computer operation. It can also represent the structural information of RNA completely and ensure that the structural information of RNA is not easily lost. The dynamic programming algorithm is used to calculate the comparison score between structures, and the heuristic algorithm is used to calculate the optimal comparison between multibranch rings. Finally, the weights are adjusted by using the neural network algorithm and the optimal ratio value and comparison structures are obtained. In comparison with the existing SimTree algorithm through experimental analysis, the proposed method showed higher accuracy and could ensure that the structural information is not easily lost in terms of improved specificity, sensitivity, and the Matthews correlation coefficient.