Accurate identification of polyadenylation sites from 30 end deep sequencing using a na?̈ve Bayes classifier

Abstract

Motivation: 30 end processing is important for transcription termination, mRNA stability and regulation of gene expression. To identify 30 ends, most techniques use an oligo-dT primer to construct deep sequencing libraries. However, this approach can lead to identification of artifactual polyadenylation sites due to internal priming in homopolymeric stretches of adenines. Although heuristic filters have been applied in these cases, they typically result in a high proportion of both false-positive and -negative classifications. Therefore, there is a need to develop improved algorithms to better identify mis-priming events in oligo-dT primed sequences. Results: By analyzing sequence features flanking 30 ends derived from oligo-dT-based sequencing, we developed a na?̈ve Bayes classifier to classify them as true or false/internally primed. The resulting algorithm is highly accurate, outperforms previous heuristic filters and facilitates identification of novel polyadenylation sites. Contact: nathan.lawson@umassmed.edu Supplementary information: Supplementary data are available at Bioinformatics online. © The Author 2013. Published by Oxford University Press.