A Deep Learning Model for MicroRNA-Target Binding Ahmet Paker and Hasan O˘ gul Abstract MicroRNAs (miRNAs) are non-coding RNAs of ~21–23 bases length, which play critical role in gene expression. They bind the target mRNAs in the post-transcriptional level and cause translational inhibition or mRNA cleavage. Quick and effective detection of the binding sites of miRNAs is a major problem in bioinformatics. This chapter introduces a new technique to model microRNA-target binding using Recurrent Neural Networks (RNN) over a miRNA-target duplex sequence representation. Keywords Deep Learning · Recurrent Neural Networks · Long-Short Term Memory · Sequence Alignment · miRNA · target prediction · miRNA target site 1 Introduction MicroRNAs (miRNAs) are small and non-coding RNA molecules of ~21–23 bases length, which play an important role in gene expression. After transcription, they bind to target mRNAs and cause mRNA cleavage or translation inhibition in many living organisms. They bind their partial complementary target site and cause cleavage or posttranscriptional repression. They prohibit the genesis of peptides and output proteins [1, 2]. Recent research shown that gene regulation of psychiatric and neurodevelopmental disorders can be observable because of some miRNAs [3]. Since their function is usually elucidated and interpreted by the activities of their target mRNA molecules, rapid and efficient determination of the binding sites of A. Paker Department of Computer Engineering, Ba¸ skent University, Ankara, Turkey H. O˘ gul () Faculty of Computer Sciences, Østfold University College, Halden, Norway e-mail: hasan.ogul@hiof.no © Springer Nature Switzerland AG 2021 M. Elloumi (ed.), Deep Learning for Biomedical Data Analysis, https://doi.org/10.1007/978-3-030-71676-9_3 61