Interpretable detection of novel human viruses from genome sequencing data

Abstract

Viruses evolve extremely quickly, so reliable meth- ods for viral host prediction are necessary to safe- guard biosecurity and biosafety alike. Novel human- infecting viruses are difficult to detect with stan- dard bioinformatics workflows. Here, we predict whether a virus can infect humans directly from next- generation sequencing reads. We show that deep neural architectures significantly outperform both shallow machine learning and standard, homology- based algorithms, cutting the error rates in half and generalizing to taxonomic units distant from those presented during training. Further, we develop a suite of interpretability tools and show that it can be applied also to other models beyond the host pre- diction task. We propose a new approach for con- volutional filter visualization to disentangle the in- formation content of each nucleotide from its contri- bution to the final classification decision. Nucleotide- resolution maps of the learned associations between pathogen genomes and the infectious phenotype can be used to detect regions of interest in novel agents, for example, the SARS-CoV-2 coronavirus, unknown before it caused a COVID-19 pandemic in 2020. All methods presented here are implemented as easy- to-install packages not only enabling analysis of NGS datasets without requiring any deep learning skills, but also allowing advanced users to easily train and explain new models for genomics.