A systematic evaluation of computational methods for predicting translated non-canonical ORFs from ribosome profiling data

Ribosome profiling(Ribo-Seq)enables genome-wide survey of translational activities at nucleotide resolution by enriching and sequencing ribosome-protected fragments(RPFs)in the cell(Ingolia et al.,2009).Based on the characteristic distribution of RPFs in Ribo-Seq,such as enrichment in open reading frames(ORFs)and three-nucleotide periodicity,many computational methods have been developed to detect translated ORFs from Ribo-Seq data(Wang et al.,2020a).Their widespread application leads to the identification of tens of thousands of non-canonical ORFs(ncORFs)in untranslated regions(UTRs)of mRNAs and long non-coding RNAs(IncRNAs)in different species(Li et al.,2021).