Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Taxonomy01:31

Taxonomy

84.2K
Taxonomy is the science of defining and naming groups of biological organisms based on shared characteristics. It uses a hierarchy of increasingly inclusive categories with Latin names. The smallest units of taxonomy, species and genus, are used to assign a formal, taxonomic name to each species in a system. This classification system, referred to as binomial nomenclature, was formalized by Carolus Linnaeus in the 18th century.
Hierarchy of Taxonomy
The hierarchy that Carolus Linnaeus first...
84.2K
Genetic Lingo01:11

Genetic Lingo

113.6K
Overview
113.6K
The Tree of Life - Bacteria, Archaea, Eukaryotes02:40

The Tree of Life - Bacteria, Archaea, Eukaryotes

37.5K
The “tree of life” describes the evolution of life and the evolutionary relationships between organisms. The root of the tree is the common ancestor to all life on Earth. All other species radiate from this point, much like the branches of a tree. The numerous tips of these branches on the tree of life represent every living, or extant, species. Extinct species, which are species that no longer exist, can be found towards the center of the tree. Currently, these organisms, both...
37.5K
Phylogenetic Trees03:21

Phylogenetic Trees

49.1K
Phylogenetic trees come in many forms. It matters in which sequence the organisms are arranged from the bottom to the top of the tree, but the branches can rotate at their nodes without altering the information. The lines connecting individual nodes can be straight, angled, or even curved.
49.1K
Modern Molecular Taxonomy01:29

Modern Molecular Taxonomy

554
Advancements in molecular biology have revolutionized the identification and characterization of bacteria, with multiple methods leveraging DNA sequencing for enhanced precision. As sequencing technologies improve and costs decline, these approaches are increasingly used in clinical, environmental, and evolutionary studies.Multilocus Sequence Typing (MLST) examines several housekeeping genes, essential chromosomal genes encoding cellular functions, to distinguish strains. Approximately...
554
Applications of Molecular Taxonomy01:20

Applications of Molecular Taxonomy

489
Molecular taxonomy has revolutionized the understanding and classification of bacteria, providing precise insights into their diversity, evolutionary relationships, and ecological roles. By utilizing molecular techniques such as DNA sequencing and fingerprinting, researchers have made significant strides in various fields related to bacterial studies.Resolving Taxonomic AmbiguitiesMolecular taxonomy has been instrumental in distinguishing closely related bacterial species initially thought to...
489

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Protein language models are accidental taxonomists.

BMC bioinformatics·2026
Same author

A microphysiologic human cervical model recapitulates microbial, immune, and pathogenic properties of sexually transmitted infections.

Science advances·2026
Same author

BEADS: An Interactive Semi-Automated Workflow for 3D Fibrin Angiogenesis Assays Enabling Co-Culture and Directionality Analysis.

bioRxiv : the preprint server for biology·2025
Same author

Modular Parallel Plate Flow Chamber with Tunable Substrate Mechanics and Defined Shear Stress.

bioRxiv : the preprint server for biology·2025
Same author

Anatomical 3D Reconstruction of Murine Lymph Nodes for Visualization, Quantitation, and Numerical Simulation.

bioRxiv : the preprint server for biology·2025
Same author

Confocal-Compatible Workflow for Sectioning, Staining, and Imaging Serial Vibratome Sections for 3D Anatomical Reconstruction of the Lymph Node.

bioRxiv : the preprint server for biology·2025

Related Experiment Video

Updated: Jan 10, 2026

A Concoction Pipeline for Generating Molecular Operational Taxonomic Units (MOTUs) Among Riparian and Aquatic Beetles
10:23

A Concoction Pipeline for Generating Molecular Operational Taxonomic Units (MOTUs) Among Riparian and Aquatic Beetles

Published on: July 11, 2025

554

Protein Language Models are Accidental Taxonomists.

Logan Hallee1,2, Tamar Peleg3, Nikolaos Rafailidis1

  • 1Center for Bioinformatics and Computational Biology, University of Delaware.

Biorxiv : the Preprint Server for Biology
|November 24, 2025
PubMed
Summary
This summary is machine-generated.

Protein language models (pLMs) can exploit species differences in protein-protein interaction (PPI) datasets, leading to inflated performance. Careful data curation is crucial for reliable computational biology predictions.

Keywords:
ConfoundersNegative samplingPhylogeneticsProtein Language ModelingProtein-Protein InteractionsTaxonomy

More Related Videos

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

Related Experiment Videos

Last Updated: Jan 10, 2026

A Concoction Pipeline for Generating Molecular Operational Taxonomic Units (MOTUs) Among Riparian and Aquatic Beetles
10:23

A Concoction Pipeline for Generating Molecular Operational Taxonomic Units (MOTUs) Among Riparian and Aquatic Beetles

Published on: July 11, 2025

554
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Machine Learning in Biology

Background:

  • Protein-protein interactions (PPIs) are vital for biological functions but experimentally challenging to identify.
  • Computational methods, especially protein language models (pLMs), offer scalable solutions for PPI prediction.
  • High performance of pLMs on multi-species datasets raises questions about the underlying prediction mechanisms.

Purpose of the Study:

  • To introduce and validate the "accidental taxonomist" hypothesis for PPI prediction.
  • To investigate whether pLMs leverage phylogenetic information instead of true interaction features.
  • To propose strategies for mitigating this confounding factor in multi-species datasets.

Main Methods:

  • Analysis of protein-protein interaction (PPI) datasets with random negative sampling.
  • Utilizing protein language model (pLM) embeddings to assess taxonomic relatedness.
  • Implementing a strategic sampling strategy for negative examples within the same species.
  • Comparing model performance across different data sampling strategies.

Main Results:

  • Real PPIs in datasets predominantly involve proteins from the same species, while negative samples often come from different species.
  • pLM embeddings can accurately predict the taxonomic origin of protein pairs, indicating reliance on phylogenetic signals.
  • Restricting negative samples to same-species pairs significantly reduces model performance, confirming the accidental taxonomist effect.
  • Strategically curated multi-species models outperform single-species models, highlighting the potential of well-managed data.

Conclusions:

  • The "accidental taxonomist" is a significant confounder in multi-species PPI prediction, where models exploit species labels rather than biological interactions.
  • This phylogenetic bias is not exclusive to PPIs and may affect other supervised learning tasks in computational biology.
  • Careful data curation and sampling strategies are essential to harness the benefits of multi-species data for accurate and reliable computational predictions.