Emma Bohn

DeepRNAi predicts siRNA on- and off-target activity using deep learning, showing high accuracy across species and genomes, and highlights off-target concerns in potent siRNAs.

This paper introduces REPRESS, a deep learning model that predicts cell-type-specific microRNA binding and mRNA degradation from RNA sequences, outperforming existing methods in multiple tasks.

The paper curates a high-confidence set of pathogenic and likely pathogenic UTR variants and evaluates deep learning models for predicting their functional impact.