David Heurtel--Depeiges

david.heurtel-depeiges@mila.quebec

PhD - Polytechnique Montréal

Supervisor

Sarath Chandar

Research Topics

AI for Science

AI Safety

Natural Language Processing

Website

Google Scholar

GitHub

Publications

pLM representations unlock metagenomic space beyond homology

Lola Le Breton

David Heurtel-Depeiges

Douglas C. Millar

Lara E. Zetzsche

Robert M. Vernon

Christopher J. Langmead

Sarath Chandar

Quentin Fournier

Metagenomic sequencing has uncovered billions of proteins from uncultured microorganisms, vastly expanding the known protein space. Yet most… (see more) remain functionally inaccessible because existing annotation methods depend on close homologs or accurate structure predictions. Here, we show that protein language models (pLMs) can unlock this diversity only when their training data are appropriately curated. We introduce Residue Embedding Diversity (RED), a metric for protein quality assessment orders of magnitude cheaper than likelihood, and a calibration task that measures model alignment with natural evolutionary distributions. We discover a fundamental trade-off between evolutionary calibration and structural modeling, establishing training data composition as a primary determinant of pLM behavior. Finally, we successfully retrieve diverse enzyme candidates from billions of metagenomic sequences and validate their expression in vivo .

2026-07-29

bioRxiv (preprint)

doi.org

Structure-Aligned Protein Language Model

Can Chen

David Heurtel-Depeiges

Robert M. Vernon

Christopher J. Langmead

Yoshua Bengio

Quentin Fournier

2025-05-21

ArXiv (preprint)