torchtree : Flexible Phylogenetic Model Development and Inference Using PyTorch
Systematic Biology · 15.01.2026 · 75 (1) · 39-51
DOI: 10.1093/sysbio/syaf047 ↗Kurz gefasst
Fachartikel in der Zeitschrift Systematic Biology, online erschienen am 04.07.2025. Verfasst von Mathieu Fourment und 5 weiteren Autor:innen. Beteiligte Einrichtung: Australian Institute for Microbiology and Infection, University of Technology Sydney, 5 Broadway, Ultimo, NSW 2007, Australia. Hauptthemen laut MEDLINE: Phylogeny, Classification, Software. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, NIAID NIH HHS, Louisiana Board of Regents.
Abstract (Originalsprache)
Bayesian inference has predominantly relied on the Markov chain Monte Carlo (MCMC) algorithm for many years. However, MCMC is computationally laborious, especially for complex phylogenetic models of time trees. This bottleneck has led to the search for alternatives, such as variational Bayes, which can scale better to large data sets. In this paper, we introduce torchtree, a framework written in Python that allows developers to easily implement rich phylogenetic models and algorithms using a fixed tree topology. One can either use automatic differentiation or leverage torchtree’s plug-in system to compute gradients analytically for model components for which automatic differentiation is slow. We demonstrate that the torchtree variational inference framework performs similarly to BEAST in terms of speed, and delivers promising approximation results, though accuracy varies across scenarios. Furthermore, we explore the use of the forward Kullback–Leibler (KL) divergence as an optimizing criterion for variational inference, which can handle discontinuous and nondifferentiable models. Our experiments show that inference using the forward KL divergence is frequently faster per iteration compared with the evidence lower bound (ELBO) criterion, although the ELBO-based inference may converge faster in some cases. Overall, torchtree provides a flexible and efficient framework for phylogenetic model development and inference using PyTorch.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Australian Institute for Microbiology and Infection, University of Technology Sydney, 5 Broadway, Ultimo, NSW 2007, Australia.
- Centre for Computational Evolution, The University of Auckland, 303/38 Princes St, Auckland, 1010, New Zealand.
- School of Computer Science, The University of Auckland, 303/38 Princes St, Auckland, 1010, New Zealand.
- Department of Mathematics, Tulane University, 6823 St. Charles Avenue, New Orleans, LA 70118, USA.
- Department of Human Genetics, University of California, 885 Tiverton Drive, Los Angeles, CA 90095, USA.
- Department of Computational Medicine, University of California, 885 Tiverton Drive, Los Angeles, CA 90095, USA.
ORCID-Profile: Mathieu Fourment, Matthew Macaulay, Marc A Suchard, Frederick A Matsen Iv
Förderung
- NIH HHS
- NIAID NIH HHS
- Louisiana Board of Regents
- NSF
- Australian Research Council
- Fred Hutch
- National Institutes of Health
Zitieren
APA
Fourment, M., Macaulay, M., Swanepoel, C. J., Ji, X., Suchard, M. A., & Matsen Iv, F. A. (2026). torchtree : Flexible Phylogenetic Model Development and Inference Using PyTorch. Systematic Biology, 75(1), 39-51. https://doi.org/10.1093/sysbio/syaf047
Vancouver
Fourment M, Macaulay M, Swanepoel CJ, Ji X, Suchard MA, Matsen Iv FA. torchtree : Flexible Phylogenetic Model Development and Inference Using PyTorch. Systematic Biology. 2026;75(1):39-51. doi:10.1093/sysbio/syaf047
BibTeX
@article{Fourment2026,
author = {Fourment, Mathieu and Macaulay, Matthew and Swanepoel, Christiaan J and Ji, Xiang and Suchard, Marc A and Matsen Iv, Frederick A},
title = {torchtree : Flexible Phylogenetic Model Development and Inference Using PyTorch},
journal = {Systematic Biology},
year = {2026},
volume = {75},
number = {1},
pages = {39--51},
doi = {10.1093/sysbio/syaf047}
}
EUVIMED