Inferring variant-specific effective reproduction numbers from combined case and sequencing data

Accurately estimating relative transmission rates of SARS-CoV-2 variants remains a scientific and public health priority. Recent studies have used the sample proportions of different variants from genetic sequence data to describe variant frequency dynamics and relative transmission rates, but frequencies alone cannot capture the rich epidemiological behavior of SARS-CoV-2. Here, we extend methods for inferring the effective reproduction number of an epidemic using confirmed case data to jointly estimate variant-specific effective reproduction numbers and frequencies of co-circulating variants using cases and sequences across states in the United States from January 2021 to March 2022. Our method can be used to infer structured relationships between effective reproduction numbers across time series, allowing us to estimate fixed variant-specific growth advantages. We use this model to estimate the effective reproduction number of SARS-CoV-2 variants of concern and variants of interest in the United States, and to estimate consistent growth advantages of particular variants across different locations.