Alternative polyadenylation(APA)is a crucial step in post-transcriptional regulation.Previous bioinformatic studies have mainly focused on the recognition of polyadenylation sites(PASs)in a given genomic sequence,which is a binary classification problem.Recently,computa-tional methods for predicting the usage level of alternative PASs in the same gene have been pro-posed.However,all of them cast the problem as a non-quantitative pairwise comparison task and do not take the competition among multiple PASs into account.To address this,here we propose a deep learning architecture,Deep Regulatory Code and Tools for Alternative Polyadenylation(DeeReCT-APA),to quantitatively predict the usage of all alternative PASs of a given gene.To accommodate different genes with potentially different numbers of PASs,DeeReCT-APA treats the problem as a regression task with a variable-length target.Based on a convolutional neural network-long short-term memory(CNN-LSTM)architecture,DeeReCT-APA extracts sequence features with CNN layers,uses bidirectional LSTM to explicitly model the interactions among com-peting PASs,and outputs percentage scores representing the usage levels of all PASs of a gene.In addition to the fact that only our method can quantitatively predict the usage of all the PASs within a gene,we show that our method consistently outperforms other existing methods on three different tasks for which they are trained:pairwise comparison task,highest usage prediction task,and rank-ing task.Finally,we demonstrate that our method can be used to predict the effect of genetic variations on APA patterns and sheds light on future mechanistic understanding in APA regulation.

Zhongxiao Li;Yisheng Li;Bin Zhang;Yu Li;Yongkang Long;Juexiao Zhou;Xudong Zou;Min Zhang;Yuhui Hu;Wei Chen;Xin Gao

King Abdullah University of Science and Technology(KAUST),Computational Bioscience Research Center(CBRC),Computer,Electrical and Mathematical Sciences and Engineering(CEMSE)Division,Thuwal 23955-6900,Saudi Arabia;Department of Biology,Southern University of Science and Technology(SUSTech),Shenzhen 518055,China;Cancer Science Institute of Singapore,Singapore 117599,Singapore

基因组蛋白质组与生物信息学报（英文版）

[1]Zhongxiao Li;Yisheng Li;Bin Zhang;Yu Li;Yongkang Long;Juexiao Zhou;Xudong Zou;Min Zhang;Yuhui Hu;Wei Chen;Xin Gao-.DeeReCT-APA:Prediction of Alternative Polyadenylation Site Usage Through Deep Learning)[J].基因组蛋白质组与生物信息学报（英文版）,2022(03):483-495

DeeReCT,Polyadenylation,PASs,peting

APA,Prediction,Alternative,Site,Usage,Through,Deep,Learning,polyadenylation,crucial,step,post,transcriptional,regulation,Previous,bioinformatic,studies,have,mainly,focused,recognition,sites,given,genomic,sequence,which,binary,classification,problem,Recently,computa,methods,predicting,usage,alternative,same,been,posed,However,them,cast,pairwise,comparison,do,not,take,competition,among,multiple,into,account,address,this,here,propose,deep,learning,architecture,Regulatory,Code,Tools,quantitatively,accommodate,different,genes,potentially,numbers,treats,regression,variable,length,target,Based,convolutional,neural,network,long,short,term,memory,extracts,features,layers,uses,bidirectional,explicitly,model,interactions,outputs,percentage,scores,representing,levels,In,addition,fact,that,only,our,can,within,show,consistently,outperforms,other,existing,three,tasks,they,are,trained,highest,prediction,rank,Finally,demonstrate,effect,genetic,variations,patterns,sheds,light,future,mechanistic,understanding

0.49967