N6-methyladenosine(m6A)is a prevalent methyla-tion modification and plays a vital role in various biological processes,such as metabolism,mRNA processing,synthesis,and transport.Recent studies have suggested that m6A modifi-cation is related to common diseases such as cancer,tumours,and obesity.Therefore,accurate prediction of methylation sites in RNA sequences has emerged as a critical issue in the area of bioinformatics.However,traditional high-throughput sequen-cing and wet bench experimental techniques have the disadvan-tages of high costs,significant time requirements and inac-curate identification of sites.But through the use of traditional experimental methods,researchers have produced many large databases of m6A sites.With the support of these basic data-bases and existing deep learning methods,we developed an m6A site predictor named DeepM6ASeq-EL,which integrates an ensemble of five LSTM and CNN classifiers with the com-bined strategy of hard voting.Compared to the state-of-the-art prediction method WHISTLE(average AUC 0.948 and 0.880),the DeepM6ASeq-EL had a lower accuracy in m6A site prediction(average AUC:0.861 for the full transcript models and 0.809 for the mature messenger RNA models)when tested on six independent datasets.

Juntao CHEN;Quan ZOU;Jing LI

Institute of Fundamental and Frontier Sciences,University of Electronic Science and Technology of China,Chengdu 610051,China;Hainan Key Laboratory for Computational Science and Application,Hainan Normal University,Haikou 571158,China;College of Intelligence and Computing,Tianjin University,Tianjin 300350,China

计算机科学前沿

[1]Juntao CHEN;Quan ZOU;Jing LI-.DeepM6ASeq-EL:prediction of human N6-methyladenosine(m6A)sites with LSTM and ensemble learning)[J].计算机科学前沿,2022(02):25-31

DeepM6ASeq,methyla,WHISTLE

EL,prediction,human,N6,methyladenosine,m6A,sites,ensemble,learning,prevalent,modification,plays,vital,role,various,biological,processes,such,metabolism,processing,synthesis,transport,Recent,studies,have,suggested,that,related,common,diseases,cancer,tumours,obesity,Therefore,accurate,methylation,sequences,has,emerged,critical,issue,area,bioinformatics,However,traditional,high,throughput,cing,wet,bench,experimental,techniques,disadvan,tages,costs,significant,requirements,inac,identification,But,use,methods,researchers,produced,many,large,databases,With,support,these,basic,existing,deep,developed,predictor,named,which,integrates,five,classifiers,bined,strategy,hard,voting,Compared,state,art,average,had,lower,accuracy,full,transcript,models,mature,messenger,when,tested,six,independent,datasets

0.568291