Genetic Diversity Drives the Rate and Fitness Jumps of Detectable SARS-CoV-2 Recombination

Viral recombination is widely considered a potent driver of adaptation, yet the epidemiological factors that influence it and the extent of its contribution to overall fitness gains of a virus remain poorly understood. In this study, we leverage three years of extensive genomic and epidemiological data collected globally for SARS-CoV-2 during the COVID-19 pandemic, combined with large-scale, unbiased recombination inference methods, to investigate recombination through an epidemiological lens. Using over 2,000 recombination events inferred from 16 million SARS-CoV-2 genomes, we show that the rate of detectable recombination is driven primarily by the standing viral genetic diversity in the population, and, to a lesser extent, by the number of infections. Our analysis highlights that >80% of detectable recombination events produce viruses with neutral or reduced fitness, indicating that recombination is consistent with the "nearly neutral theory of molecular evolution". We also find that recombination induces larger shifts in fitness--both positive and negative--than single-nucleotide substitutions, particularly when recombinants arise from genetically divergent parental lineages. We observe that recombination between genetically divergent parents can lead to two types of epidemiological consequences. In very rare instances, <0.8% of all recombination events, it can lead to highly transmissible variants with substantial and immediate fitness gains. Additionally, seemingly neutral fitness effects arising from recombination between divergent parents, which arise more frequently (12.4% of all recombinants), may pose a subtler risk by discovering novel areas within the fitness landscape, thereby creating new pathways for rapid adaptation. These findings not only deepen our understanding of viral recombination but also have important implications for genomic surveillance efforts during a pandemic, such as to identify the high-risk periods for the emergence of recombinant variants and to guide interventions.