From sequences to therapeutics: Machine learning predicts chemically modified siRNA activity

Abstract

AO_SCPLOWBSTRACTC_SCPLOWSmall interfering RNAs (siRNAs) exemplify the promise of genetic medicine in the discovery of novel therapeutic modalities. Their ability to selectively suppress gene expression makes them ideal candidates for development as oligonucleotide pharmaceuticals. Recent advancements in machine learning (ML) have facilitated unmodified siRNA design and efficacy prediction, but a model trained to predict the silencing activity of siRNAs with diverse chemical modification patterns has yet to be published, despite the importance of such chemical modifications in designing siRNAs with the potential to advance to the clinic. This study presents the first application of ML to classify efficient chemically modified siRNAs from sequence and chemical modification patterns alone. Three algorithms are evaluated at three classification thresholds and compared according to sensitivity, specificity, consistency of feature weights with empirical knowledge, and performance on an external validation dataset. Finally, possible directions for future research are proposed.