Learning the PTM Code through a Coarse-to-Fine, Mechanism-Aware Framework

2025-10-27

AI Summary

Deciphering the mapping between post-translational modification (PTM) sites and their catalyzing enzymes—termed the PTM “combinatorial code”—remains a central challenge in understanding cellular signaling regulation and disease mechanisms. This work introduces the first mechanism-aware, coarse-to-fine unified framework that jointly models multi-label PTM site prediction and zero-shot enzyme identification, explicitly encoding synergistic or antagonistic syntactic relationships among PTMs. The method integrates evolution-informed protein language model representations, physicochemical priors, and interaction-aware prompting to effectively mitigate the dual long-tail distribution inherent in PTM data. Evaluated on multiple proteome-scale benchmarks, our approach achieves a 122% improvement in site-level F1 score and a 54% gain in zero-shot enzyme assignment accuracy. Moreover, it successfully identifies PTM rewiring events triggered by disease-associated variants.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Post-translational modifications (PTMs) form a combinatorial "code" that regulates protein function, yet deciphering this code - linking modified sites to their catalytic enzymes - remains a central unsolved problem in understanding cellular signaling and disease. We introduce COMPASS-PTM, a mechanism-aware, coarse-to-fine learning framework that unifies residue-level PTM profiling with enzyme-substrate assignment. COMPASS-PTM integrates evolutionary representations from protein language models with physicochemical priors and a crosstalk-aware prompting mechanism that explicitly models inter-PTM dependencies. This design allows the model to learn biologically coherent patterns of cooperative and antagonistic modifications while addressing the dual long-tail distribution of PTM data. Across multiple proteome-scale benchmarks, COMPASS-PTM establishes new state-of-the-art performance, including a 122% relative F1 improvement in multi-label site prediction and a 54% gain in zero-shot enzyme assignment. Beyond accuracy, the model demonstrates interpretable generalization, recovering canonical kinase motifs and predicting disease-associated PTM rewiring caused by missense variants. By bridging statistical learning with biochemical mechanism, COMPASS-PTM unifies site-level and enzyme-level prediction into a single framework that learns the grammar underlying protein regulation and signaling.

Problem

• Deciphering the PTM code linking modified sites to enzymes
• Modeling inter-PTM dependencies and long-tail data distributions
• Unifying site-level and enzyme-level prediction into one framework

Innovation

• Coarse-to-fine framework integrating residue profiling and enzyme assignment
• Protein language models combined with physicochemical priors and crosstalk
• Addresses long-tail PTM data distribution for interpretable generalization