A novel computational approach to reconstruct SARS-CoV-2 infection dynamics through the inference of unsampled sources of infection

ABSTRACT

Infectious diseases such as the COVID19 pandemic cemented the importance of disease tracking. The role of asymptomatic, undiagnosed individuals in driving infection has become evident. Their unaccountability results in ineffective prevention. We developed a pipeline using genomic data to accurately predict a populations transmission network complete with the inference of unsampled sources. The system utilises Bayesian phylogenetics to capture evolutionary and infection dynamics of SARS-CoV-2. It identified the effectiveness of preventive measures in Canadas Atlantic bubble and mobile populations such as New York State. Its robustness extends to the prediction of cross-species disease transmission as we inferred SARS-CoV-2 transmission from humans to lions and tigers in New York Citys Bronx Zoo. The proposed methods ability to generate such complete transmission networks, provides a more detailed insight into the transmission dynamics within a population. This potential frontline tool will be of direct help in "the battle to bend the curve".