Preprint describes direct estimation of genotype fitness from time-series population data

A preprint posted to bioRxiv presents a new analytical framework for estimating the fitness of individual genotypes directly from time-series data in heterogeneous, adapting populations. The authors address the challenge that standard fitness inference methods typically require assumptions about the fitness landscape — particularly the maximum order of epistasis — or depend on computationally intensive iterative optimisation algorithms.

The proposed method is applicable to contexts ranging from laboratory evolution experiments to large-scale pathogen surveillance, where many genetic lineages compete simultaneously and data are noisy. By circumventing restrictive epistasis assumptions, the approach may allow more robust fitness inference in complex, real-world evolutionary settings.

The framework is relevant to researchers in population genetics, evolutionary genomics, and infectious disease genomics, particularly those working on variant fitness in rapidly evolving pathogen populations. It also has educational relevance as an illustration of how selection, mutation, and stochastic forces interact in adapting populations. As a preprint, the work has not yet been peer-reviewed.