tRFTars: predicting the targets of tRNA-derived fragments

Abstract Background tRNA-derived fragments (tRFs) are 14–40-nucleotide-long, small non-coding RNAs derived from specific tRNA cleavage events with key regulatory functions in many biological processes. Many studies have shown that tRFs associated Argonaute (AGO) complexes and inhibit gene expression the same manner as miRNAs. However, there currently no tools for accurately predicting tRF target genes. Methods We used tRF-mRNA pairs identified by crosslinking, ligation, sequencing of hybrids (CLASH) covalent ligation endogenous AGO-bound (CLEAR)-CLIP to assess features may participate targeting, including sequence context each site interactions. applied genetic algorithm (GA) select support vector machine (SVM) construct prediction models. Results first globally influenced targeting. Among these features, most significant were minimum free folding energy (MFE), position 8 match, number bases paired duplex, length tRF, which consistent previous findings. Our constructed model yielded an area under receiver operating characteristic (ROC) curve (AUC) = 0.980 (0.977–0.983) training process AUC 0.847 (0.83–0.861) test process. The was all sites perfect Watson–Crick complementarity seed 3? untranslated region (3?-UTR) human genome. Seven nine target/nontarget genes confirmed reporter assay predicted. also validated predictions via quantitative real-time PCR (qRT-PCR). Thirteen potential top significantly down-regulated at mRNA levels overexpression (tRF-3001a, tRF-3003a or tRF-3009a). Conclusions Predictions can be obtained online, tRFTars, freely available http://trftars.cmuzhenninglab.org:3838/tar/ , is tool predict targets humans a user-friendly interface.