ROSE: a deep learning based framework for predicting ribosome stalling

Abstract

We present a deep learning based framework, called ROSE, to accurately predict ribosome stalling events in translation elongation from coding sequences based on high-throughput ribosome profiling data. Our validation results demonstrate the superior performance of ROSE over conventional prediction models. ROSE provides an effective index to estimate the likelihood of translational pausing at codon resolution and understand diverse putative regulatory factors of ribosome stalling. Also, the ribosome stalling landscape computed by ROSE can recover the functional interplay between ribosome stalling and cotranslational events in protein biogenesis, including protein targeting by the signal recognition particle (SRP) and protein secondary structure formation.