EUVIMEDEUVIMED entdecken
Open Access

Trait-mediated speciation and human-driven extinctions in proboscideans revealed by unsupervised Bayesian neural networks

Torsten Hauffe iD, Juan L Cantalapiedra iD, Daniele Silvestro iD

Science Advances · 2024

DOI: 10.1126/sciadv.adl2643 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Advances, 2024. Verfasst von Torsten Hauffe und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biology, University of Fribourg and Swiss Institute of Bioinformatics, 1700 Fribourg, Switzerland. Hauptthemen laut MEDLINE: Extinction, Biological, Bayes Theorem, Fossils, Neural Networks, Computer. Die Arbeit wurde bisher 26-mal zitiert (Crossref) und nennt 118 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Species life-history traits, paleoenvironment, and biotic interactions likely influence speciation and extinction rates, affecting species richness over time. Birth-death models inferring the impact of these factors typically assume monotonic relationships between single predictors and rates, limiting our ability to assess more complex effects and their relative importance and interaction. We introduce a Bayesian birth-death model using unsupervised neural networks to explore multifactorial and nonlinear effects on speciation and extinction rates using fossil data. It infers lineage- and time-specific rates and disentangles predictor effects and importance through explainable artificial intelligence techniques. Analysis of the proboscidean fossil record revealed speciation rates shaped by dietary flexibility and biogeographic events. The emergence of modern humans escalated extinction rates, causing recent diversity decline, while regional climate had a lesser impact. Our model paves the way for an improved understanding of the intricate dynamics shaping clade diversification.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Extinction, BiologicalBayes TheoremFossilsNeural Networks, ComputerGenetic SpeciationHumansAnimalsBiodiversity

Autor:innen und Institutionen

  • Department of Biology, University of Fribourg and Swiss Institute of Bioinformatics, 1700 Fribourg, Switzerland.
  • Departamento de Paleobiología, Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Científicas, 28006 Madrid, Spain.
  • GloCEE Global Change Ecology and Evolution Research Group, Departamento de Ciencias de la Vida, Universidad de Alcalá, 28801 Alcalá de Henares, Spain.
  • Museum für Naturkunde, Leibniz Institute for Evolution and Biodiversity Science, Invalidenstraße 43, 10115 Berlin, Germany.
  • Department of Biological and Environmental Sciences, Gothenburg Global Biodiversity Centre, University of Gothenburg, 40530 Gothenburg, Sweden.

ORCID-Profile: Torsten Hauffe, Juan L Cantalapiedra, Daniele Silvestro

Zitieren

APA

Hauffe, T., Cantalapiedra, J. L., & Silvestro, D. (2024). Trait-mediated speciation and human-driven extinctions in proboscideans revealed by unsupervised Bayesian neural networks. Science Advances. https://doi.org/10.1126/sciadv.adl2643

Vancouver

Hauffe T, Cantalapiedra JL, Silvestro D. Trait-mediated speciation and human-driven extinctions in proboscideans revealed by unsupervised Bayesian neural networks. Science Advances. 2024. doi:10.1126/sciadv.adl2643

BibTeX
@article{Hauffe2024,
  author = {Hauffe, Torsten and Cantalapiedra, Juan L and Silvestro, Daniele},
  title = {Trait-mediated speciation and human-driven extinctions in proboscideans revealed by unsupervised Bayesian neural networks},
  journal = {Science Advances},
  year = {2024},
  doi = {10.1126/sciadv.adl2643}
}
Kostenlos starten →