Illuminating the Druggable Proteome with an AI Protein Profiling Platform

Creating a proteomewide atlas of reversible and covalent protein-ligand binding sites would transform our understanding of protein functions and accelerate therapeutic discovery; however, current approaches face significant challenges. Chemical proteomics is constrained by limited proteome coverage and data heterogeneity, while existing machine learning models have limited predictive power due to structural dependencies and inadequate handling of heterogeneous experimental labels. Here we developed AiPP, an AI protein profiling platform that provides comprehensive annotations of ligand interaction sites directly from sequence. AiPP is powered by the evolutionary scale protein large language models (ESM3 and ESMC) and two new databases, LigCysABPP (cysteine ligandability from activity-based protein profiling or ABPP) and LigBind3D (reversible binding sites in co-crystal structures). LigCysABPP comprises >700,000 cysteinesite records from 15 ABPP studies covering >10,000 human proteins. We developed an evolutionary representation-based clustering framework followed by consensus analysis to reconcile and augment heterogeneous experimental annotations. Two complementary iterative data expansion protocols were implemented to enhance model performance and generalization. AiPP recovers 80% (Top-1) of all cysteine liganding events from the Protein Data Bank with AUPRC of 84% and AUROC of 89%. Furthermore, it recapitulates the consistently and heterogeneously liganded cysteines determined by a recent ABPP study using over 400 cancer cell lines. AiPP provides a starting point for creating a proteome-wide atlas of protein-ligand interactions and systematic discovery of druggable sites. The integrative approach that leverages evolutionary information to harmonize largescale proteomic data can be broadly applied to protein research.