Abstract Recombination is a key molecular mechanism for the evolution and adaptation of viruses 1, 2 . The first recombinant SARS-CoV-2 genomes were recognized in 2021 3 ; as of today, more than seventy SARS-CoV-2 lineages are designated as recombinant. In the wake of the COVID-19 pandemic, several methods for detecting recombination in SARS-CoV-2 have been proposed; however, none could faithfully reproduce manual analyses by experts in the field. We hereby present RecombinHunt, a novel, automated method for the identification of recombinant genomes purely based on a data-driven approach. RecombinHunt compares favorably with other state-of-the-art methods and recognizes recombinant SARS-CoV-2 genomes (or lineages) with one or two break-points with high accuracy, within reduced turn-around times and small discrepancies with respect to the expert manually-curated standard nomenclature. Strikingly, applied to the complete collection of viral sequences from the recent monkeypox epidemic, RecombinHunt identifies recombinant viral genomes in high concordance with manually curated analyses by experts, suggesting that our approach is robust and can be applied to any epidemic/pandemic virus. Although RecombinHunt does not substitute manual expert curation based on phylogenetic analysis, we believe that our method represents a breakthrough for the detection of recombinant viral lineages in pandemic/epidemic scenarios.

Data-driven recombination detection in viral genomes