Published on Jul 30, 2025 by Arcadia Science Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2
Protein language models are trained on evolutionarily related sequences, yet the extent to which they capture the underlying evolutionary relationships remains unclear. We explore this question using reconstructed ancestral protein sequences and the ESM2 protein language model.
Contributors (A-Z)
Version 1 · Jul 30, 2025
Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2
The full pub is available here.
The source code to generate it is available in this GitHub repo (DOI: 10.5281/zenodo.16620544).
In the future, we hope to host notebook pubs directly on our publishing platform. Until that’s possible, we’ll create stubs like this with key metadata like the DOI, author roles, citation information, and an external link to the pub itself.
References
Pub details
Arcadia
York R, Nocedal I. (2025). Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2. https://doi.org/10.57844/arcadia-5cwu-spn8
apa
York, R., & Nocedal, I. (2025). Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2. Arcadia Science. https://doi.org/10.57844/arcadia-5cwu-spn8
harvard
York, R. and Nocedal, I. (2025) 'Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2', Arcadia Science [Preprint]. doi:10.57844/arcadia-5cwu-spn8.
vancouver
1.
York R, Nocedal I. Do protein language models understand evolution? Mixed evidence from ancestral sequences and ESM2. Arcadia Science [Internet]. 2025 Jul 30; Available from: https://research.arcadiascience.com/pub/result-ancestral-sequences-protein-language-models/release/1
bibtex
@article{York2025Do,
author = {York, Ryan and Nocedal, Isabel},
journal = {Arcadia Science},
doi = {10.57844/arcadia-5cwu-spn8},
issn = {2998-4084},
year = {2025},
month = {jul 30},
publisher = {Arcadia Science},
title = {Do protein language models understand evolution? {Mixed} evidence from ancestral sequences and {ESM2}},
url = {https://research.arcadiascience.com/pub/result-ancestral-sequences-protein-language-models/release/1},
}
Evan Kiefl
Validation
Isabel Nocedal
Conceptualization, Formal Analysis, Software, Visualization, Writing
