End-to-end Learning of Evolutionary Models to Find Coding Regions in Genome Alignments
1MotivationThe comparison of genomes using models of molecular evolution is a powerful approach for finding or towards understanding functional elements. In particular, comparative genomics is a fundamental building brick in building high-quality, complete and consistent annotations of ever larger sets of alignable genomes. ResultsWe here present our new program ClaMSA that classifies multiple sequence alignments using a phylogenetic model. It uses a novel continuous-time Markov chain machine learning layer, named CTMC, that is learned end-to-end together with (recurrent) neural networks for a learning task. We trained ClaMSA discriminately to classify aligned codon sequences that are candidates of coding regions into coding or non-coding and obtained six times fewer false positives for this task on vertebrate and fly alignments than existing methods at the same true positive rate. ClaMSA and the CTMC layer are general tools that could be used for other machine learning tasks on tree-related sequence data. AvailabilityFreely from https://github.com/Gaius-Augustus/clamsa.