EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

PhyloCNN: Improving Tree Representation and Neural Network Architecture for Deep Learning from Trees in Phylodynamics and Diversification Studies

Manolo Fernandez Perez iD, Olivier Gascuel iD

Systematic Biology · 12.06.2026 · 75 (4) · 776-795

DOI: 10.1093/sysbio/syaf082 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Systematic Biology, online erschienen am 22.11.2025. Verfasst von Manolo Fernandez Perez und 1 weiteren Autor:innen. Beteiligte Einrichtung: Institut de Systématique, Evolution, Biodiversité (ISYEB, UMR 7205 CNRS, Muséum National d'Histoire Naturelle, SU, EPHE & UA), CP 39, 18 rue Buffon, …. Hauptthemen laut MEDLINE: Phylogeny, Deep Learning, Classification, Neural Networks, Computer. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Paris Research Institute for Artificial Intelligence, GENCI-IDRIS, Talent Attraction Program of the Madrid Community (Spain).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phylodynamics and diversification studies using complex evolutionary models can be challenging, especially with traditional likelihood-based approaches. As an alternative, likelihood-free simulation-based approaches have been proposed due to their ability to incorporate complex models and scenarios. Here, we propose a new simulation-based deep learning (DL) method capable of selecting birth–death models and accurately estimating their parameters in both phylodynamics and diversification studies. We use a convolutional approach, where trees are encoded using the neighborhood of all nodes and leaves of the input phylogeny. We also developed a dedicated neural network architecture called PhyloCNN. Using simulations, we compared the accuracy of PhyloCNN when using a variable number of neighbors to describe the local context of nodes and leaves. The number of neighbors had a greater impact when considering smaller training sets, with a broader context showing higher accuracy, especially for complex evolutionary models. Compared with other recently developed DL approaches, PhyloCNN showed higher or similar accuracies for all parameters when used with training sets one or two orders of magnitude smaller (10,000 to 100,000 simulated training trees, instead of millions). PhyloCNN also compared favorably with state-of-the-art likelihood-based methods. We applied PhyloCNN with compelling results to two real-world phylodynamics and diversification data sets, related to HIV super-spreaders in Zurich and to primates and their ecological role as seed dispersers. The high accuracy and computational efficiency of PhyloCNN opens new possibilities for phylodynamics and diversification studies that need to account for idiosyncratic phylogenetic histories with specific parameter spaces and sampling scenarios.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PhylogenyDeep LearningClassificationNeural Networks, ComputerAnimalsComputer Simulation

Stichwörter der Autor:innen

Deep learningHIVdiversificationphylodynamicsprimatestree modelstree representations

Autor:innen und Institutionen

  • Institut de Systématique, Evolution, Biodiversité (ISYEB, UMR 7205 CNRS, Muséum National d'Histoire Naturelle, SU, EPHE & UA), CP 39, 18 rue Buffon, 75005 Paris, France.
  • Departamento de Biodiversidad y Conservación, Real Jardin Botanico-CSIC, Pl. Murillo, 2, Retiro, 28014 Madrid, Spain.

ORCID-Profile: Manolo Fernandez Perez, Olivier Gascuel

Förderung

  • Paris Research Institute for Artificial Intelligence
  • GENCI-IDRIS
  • Talent Attraction Program of the Madrid Community (Spain)
  • GENCI–IDRIS

Zitieren

APA

Perez, M. F., & Gascuel, O. (2026). PhyloCNN: Improving Tree Representation and Neural Network Architecture for Deep Learning from Trees in Phylodynamics and Diversification Studies. Systematic Biology, 75(4), 776-795. https://doi.org/10.1093/sysbio/syaf082

Vancouver

Perez MF, Gascuel O. PhyloCNN: Improving Tree Representation and Neural Network Architecture for Deep Learning from Trees in Phylodynamics and Diversification Studies. Systematic Biology. 2026;75(4):776-795. doi:10.1093/sysbio/syaf082

BibTeX
@article{Perez2026,
  author = {Perez, Manolo Fernandez and Gascuel, Olivier},
  title = {PhyloCNN: Improving Tree Representation and Neural Network Architecture for Deep Learning from Trees in Phylodynamics and Diversification Studies},
  journal = {Systematic Biology},
  year = {2026},
  volume = {75},
  number = {4},
  pages = {776--795},
  doi = {10.1093/sysbio/syaf082}
}
Kostenlos starten →