Computational Modeling Of Microrna Targeting And Context Specificity

Abstract

In this dissertation, I present two studies on miRNA regulation enabled by high-throughput sequencing technologies and computational approaches. In the first study, we attempted to learn a general model for miRNA targeting principles based on AGO CLIP and CLASH data. We used discriminative learning on AGO CLIP and CLASH interactions to train a miRNA target prediction model. Our method combined two SVM classifiers, one to predict miRNA-mRNA duplexes and a second to learn AGO’s local sequence preferences and positional bias in 3’UTR isoforms. The duplex SVM model enabled the prediction of non-canonical target sites and more accurately resolved miRNA interactions from AGO CLIP data than previous methods. The binding model was trained using a multi-task strategy to learn context-specific and common AGO sequence preferences. The duplex and common AGO binding models together outperformed existing miRNA target prediction algorithms on held-out binding data. In the second study, we attempted to characterize the context specificity of miRNA-mediated regulation of target mRNAs that are co-expressed across multiple cell types. We explored transcriptome-wide targeting and gene regulation by miR-155, whose activation-induced expression plays important roles in innate and adaptive immunity. Through mapping of miR-155 targets using differential AGO iCLIP, mRNA quantification using RNA-Seq, and 3’UTR usage analysis using polyadenylation (polyA)-Seq in activated miR-155-sufficient and -deficient macrophages, dendritic cells, T and B lymphocytes, we have identified numerous miR-155 targets with cellular context specificity. While alternative cleavage and polyadenylation (ApA) contributed to differential miR-155 binding in some transcripts, a majority of identical 3’UTR isoforms were also differentially regulated, suggesting ApA-independent and cellular context-dependent miR-155-mediated gene regulation reminiscent of sequence-specific transcription factors. Our study provides a comprehensive map of miR-155’s regulatory networks in key immune cell types.