Site4Drug: Predicting Drug-Binding Target Sites with an AI Agent
2026-06-01 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed Site4Drug, a tool that helps scientists figure out the best places on a protein to target with drugs, especially for tricky membrane proteins. Unlike other methods, Site4Drug suggests what type of drug (like an antibody or a small molecule) might work best based on detailed protein features. It provides a ranked list of target sites along with helpful info about risks and evidence, making the decision easier and more reliable. Their approach aims to avoid picking spots that look good chemically but aren't actually reachable or useful biologically.
protein targetingmembrane proteinsdrug modalitybinding sitepost-translational modificationshydropathydisulfide bondssmall-molecule drugsantibodiespeptides
Authors
Taehan Kim, Sarrah Rose Mikhail Leung, Bharat Mekala, Jeongbin Park
Abstract
Selecting where to intervene on a protein (i.e., choosing a targetable site) is often a more ambiguous and failure-prone bottleneck than selecting what binds, especially for membrane proteins where accessibility, topology, and post-translational modifications (PTMs) constrain actionable regions. We present Site4Drug, a modality-aware site-finding agent that outputs a ranked list of targetable regions with explicit constraints, evidence summaries, risk flags, and a traceable decision log. Rather than requiring users to specify the drug modality upfront, Site4Drug can recommend a binding modality (e.g., antibody/peptide-like vs small-molecule) from the same evidence used for site discovery, including topology, hydropathy, PTM propensity, disulfides, domain context, and sequence. Importantly, this evidence is applied consistently across modalities, including small-molecule pocket discovery, to avoid selecting chemically plausible but biologically occluded sites.