Protein Language Model Embeddings Reveal Proteome-Scale Ortholog Divergence Relevant to Cross-Species Pharmacology
Proteomes · 23.07.2026 · 14 (3) · 36
DOI: 10.3390/proteomes14030036 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proteomes, online erschienen am 23.07.2026. Verfasst von Taichi Endoh und 4 weiteren Autor:innen. Beteiligte Einrichtung: MATO Corporation, Exis Building 4F, 5-26-22 Higashi-Ohi, Shinagawa-ku, Tokyo 140-0011, Japan. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Rakuno Gakuen University.
Abstract (Originalsprache)
Background
Comparative proteome analysis can reveal functional conservation and divergence among orthologous proteins, with important implications for pharmacology and toxicology. Protein language models (PLMs) may capture sequence-derived functional relationships beyond what conventional alignment metrics capture.
Methods
Orthologous proteins from Danio rerio and Danio aesculapii were compared using embeddings generated by the Evolutionary Scale Modeling 2 (ESM-2) protein language model. Reciprocal best-hit inference identified 68,971 high-confidence ortholog pairs, of which 51,086 were available for embedding-based analysis. PLM divergence was quantified using cosine distance and evaluated using length-matched and bitscore-matched random controls, Gene Ontology graph-distance analysis, and localized domain-level comparisons.
Results
Ortholog pairs showed strong global conservation, with a median PLM distance of 0.000487, whereas randomized controls exhibited substantially greater divergence. Increasing Gene Ontology graph distance broadened PLM-distance distributions, and leaf–parent comparisons demonstrated significant functional ordering (Wilcoxon p = 2.44 × 10−4). Local analyses revealed increased divergence in pathophysiologically relevant regions of aryl hydrocarbon receptor (AHR) and potassium channel proteins.
Conclusions
PLM embeddings provide a scalable framework for comparative proteome characterization, complement conventional sequence-based analyses, and prioritize orthologs or protein regions with elevated functional divergence for experimental validation in cross-species pharmacology, toxicology, and systems biology.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- MATO Corporation, Exis Building 4F, 5-26-22 Higashi-Ohi, Shinagawa-ku, Tokyo 140-0011, Japan.
- School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu 069-8501, Japan.
- College of Veterinary Medicine, University of Eastern Philippines, Catarman 6400, Philippines.
- College of Science, Polytechnic University of the Philippines, Manila 1016, Philippines.
- Medical Corporation Tetsuryukai, Kasai Home Care Clinic, 101, 18-12 Nakakasai 8-chome, Edogawa-ku, Tokyo 134-0083, Japan.
- Knowledge Connect, K2 SQUARE, 7F, 12-5 Minami 2-jo Nishi 3-chome, Chuo-ku, Sapporo 060-0062, Japan.
ORCID-Profile: Gerry Amor Camer, Kotetsu Kayama, Daiji Endoh, Hiroki Teraoka
Förderung
- Rakuno Gakuen University
Zitieren
APA
Endoh, T., Camer, G. A., Kayama, K., Endoh, D., & Teraoka, H. (2026). Protein Language Model Embeddings Reveal Proteome-Scale Ortholog Divergence Relevant to Cross-Species Pharmacology. Proteomes, 14(3), 36. https://doi.org/10.3390/proteomes14030036
Vancouver
Endoh T, Camer GA, Kayama K, Endoh D, Teraoka H. Protein Language Model Embeddings Reveal Proteome-Scale Ortholog Divergence Relevant to Cross-Species Pharmacology. Proteomes. 2026;14(3):36. doi:10.3390/proteomes14030036
BibTeX
@article{Endoh2026,
author = {Endoh, Taichi and Camer, Gerry Amor and Kayama, Kotetsu and Endoh, Daiji and Teraoka, Hiroki},
title = {Protein Language Model Embeddings Reveal Proteome-Scale Ortholog Divergence Relevant to Cross-Species Pharmacology},
journal = {Proteomes},
year = {2026},
volume = {14},
number = {3},
pages = {36},
doi = {10.3390/proteomes14030036}
}
EUVIMED