Abstract
BACKGROUND: The advent of next-generation and long-read sequencing technologies has provided an ever-increasing wealth of phylogenetic data that require specially designed algorithms to decipher the underlying evolutionary relationships. As large-scale data become increasingly accessible, there is a concomitant need for efficient computational libraries that facilitate the development and dissemination of specialized algorithms for phylogenetic comparative biology. RESULTS: We introduce Phylo-rs: a fast, extensible, general-purpose library for phylogenetic analysis and inference written in the Rust programming language. Phylo-rs leverages a combination of speed, memory-safety, and native WebAssembly support offered by Rust to provide a robust set of memory-efficient data structures and elementary phylogenetic algorithms. Phylo-rs focuses on the efficient and convenient deployment of software aimed at large-scale phylogenetic analysis and inference. Scalability analysis against popular libraries shows that Phylo-rs performs comparably or better on key algorithms. We utilized it to assess the phylogenetic diversity of influenza A virus in swine, identifying virus groups that are undergoing evolutionary expansion that could be targeted for control through multivalent vaccines. Additionally, we used Phylo-rs to enhance phylogenetic inference by visualizing tree space from Markov chain Monte Carlo (MCMC) Bayesian analysis, efficiently computing approximately five billion tree pair distances to evaluate convergence and select MCMC runs for genomic epidemiology. CONCLUSION: Phylo-rs enables the design and implementation of cutting-edge software for phylogenetic analysis, thereby facilitating the application and dissemination of theoretical advancements in biology. Phylo-rs is available under an open-source license on GitHub at https://github.com/sriram98v/phylo-rs , with documentation available at https://docs.rs/phylo/latest/phylo/ .