Bi-TEAM: A Unified Cross-Scale Representation Learning Framework for Chemically Modified Biomolecules

TL;DR

Bi-TEAM is a unified framework that combines protein and chemical language models to improve representation learning for chemically modified biomolecules, enhancing accuracy in biochemical predictions and peptide design.

Contribution

It introduces a novel bi-gated residual fusion mechanism that integrates chemical details into global protein contexts without expanding the alphabet.

Findings

Achieved up to 66% improvement in MCC on scaffold-similarity splits.

Increased hemolysis prediction accuracy by 350%.

Quadrupled success rate in designing cell-penetrating cyclic peptides.

Abstract

Representation learning for protein biochemical space faces a difficult trade-off: protein language models excel at capturing long-range biological semantics but often miss fine-grained chemical details. Conversely, chemical language models encode atomic information but lack broader sequence context. To address this, we introduce Bi-TEAM (Bi-gated Residual Space Modification), a general framework that injects localized chemical variation into global protein contexts. By ensuring robustness against perturbations such as non-canonical amino acids, post-translational modifications (PTMs), and topological constraints, Bi-TEAM uncovers functional chemical dependencies often missed by evolutionary baselines. Mechanistically, Bi-TEAM maps non-canonical residues to their natural counterparts and injects atomic-level data via a bi-gated residual fusion mechanism. Crucially, this process uses modification-aware prompts to ensure that local structural changes influence global functional representations without requiring alphabet expansion. We evaluated Bi-TEAM on ten datasets spanning chemically modified peptides, PTMs, and natural proteins. The model consistently outperformed state-of-the-art baselines, achieving up to a 66 percent improvement in Matthews correlation coefficient (MCC) on scaffold-similarity splits and a 350 percent increase in hemolysis prediction accuracy. Furthermore, when deployed as an oracle for generative modeling, Bi-TEAM nearly quadrupled the success rate for designing cell-penetrating cyclic peptides. By unifying biological semantics with chemical precision, Bi-TEAM provides a versatile foundation for machine learning driven exploration of peptide and protein biochemical space.