Modeling viral mutations in the spread of epidemics

TL;DR

This paper introduces an individual-based network model for epidemic spread that explicitly incorporates viral genetic mutations, providing analytical and numerical insights into viral evolution and reinfection probabilities.

Contribution

It develops a novel model combining epidemic dynamics with explicit viral genetic mutation tracking, and validates it with SARS-CoV-2 data.

Findings

Analytical expression for average genetic distance over time.

Good agreement with SARS-CoV-2 genetic data.

Lower connectivity increases reinfection probability.

Abstract

Although traditional models of epidemic spreading focus on the number of infected, susceptible and recovered individuals, a lot of attention has been devoted to integrate epidemic models with population genetics. Here we develop an individual-based model for epidemic spreading on networks in which viruses are explicitly represented by finite chains of nucleotides that can mutate inside the host. Under the hypothesis of neutral evolution we compute analytically the average pairwise genetic distance between all infecting viruses over time. We also derive a mean-field version of this equation that can be added directly to compartmental models such as SIR or SEIR to estimate the genetic evolution. We compare our results with the inferred genetic evolution of SARS-CoV-2 at the beginning of the epidemic in China and found good agreement with the analytical solution of our model. Finally, using genetic distance as a proxy for different strains, we use numerical simulations to show that the lower the connectivity between communities, e.g., cities, the higher the probability of reinfection.