What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design
Abstract
LLM agents optimize a ligand in a protein pocket by editing it step by step and docking each edit. Recent agents first spend extra dockings probing the complex with controlled edits; guidance induced from these probes is reported to help more than the chemical knowledge the model already carries. The probes are then discarded with the instance that produced them, so the obvious next step is to keep them in an experience base, letting a new pocket inherit what earlier ones already paid for. Such a base is read through a \emph{key}: the record properties under which past edits are pooled and a new one is looked up. Using the full log of one such run over the CrossDocked2020 test-set pockets, we ask whether its keys recur on a pocket the base has never seen and still separate improving from worsening edits. \crchange{Using} the contacts at the site being edited, the action, and the round as \crchange{a} key, the base separates them sharply. Within our analysis, this separation is largely explained by how much the edit changed the molecule's size, the action, and the round, none of which describes the target. That is not because the target is irrelevant: naming it outright separates them several times more sharply. For a pocket absent from the base, transfer must rely on a description rather than its identity. Among the exact-match descriptions we evaluate, the more specific ones rarely recur, whereas the coarser ones provide little separation; the signals that do transfer primarily describe the edit and the molecule it produced rather than the target. Under the representations and lookup scheme studied here, an experience base may be more useful for prioritizing an agent's own proposals for docking than for directly prescribing what to build.
Related
- A thioester-thiol exchange reaction in Salmonella Invasion acyl carrier protein transfers the 4PP-bound acyl chain to a distal cysteine
- Metaheuristic Optimization of Temporal Surrogates for Early Prediction of Protein Conformational Stability from Short Molecular-Dynamics Trajectories
- Molecular Flexibility and Electrostatic Interactions Drive the Nanoscale Structure of Coiled-Coil GM130 Protein Condensates
- Structure-Guided Optimization of a LipL32-Targeting DNA Aptamer through Molecular Dynamics and Free-Energy Analysis
- Probe Before You Edit: Probing-Guided Molecular Optimization for LLM Agents in Structure-Based Drug Design
BibTeX
@misc{yang2026what,
title = {{What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design}},
author = {Zaifei Yang and Weiyu Chen and Yaqing Wang and James Kwok},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.32620},
eprint = {2610.32620},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.32620},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design
AU - Zaifei Yang
AU - Weiyu Chen
AU - Yaqing Wang
AU - James Kwok
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.32620
UR - https://arxiv.org/abs/2610.32620
AB - LLM agents optimize a ligand in a protein pocket by editing it step by step and docking each edit. Recent agents first spend extra dockings probing the complex with controlled edits; guidance induced from these probes is reported to help more than the chemical knowledge the model already carries. The probes are then discarded with the instance that produced them, so the obvious next step is to keep them in an experience base, letting a new pocket inherit what earlier ones already paid for. Such a base is read through a \emph{key}: the record properties under which past edits are pooled and a new one is looked up. Using the full log of one such run over the CrossDocked2020 test-set pockets, we ask whether its keys recur on a pocket the base has never seen and still separate improving from worsening edits. \crchange{Using} the contacts at the site being edited, the action, and the round as \crchange{a} key, the base separates them sharply. Within our analysis, this separation is largely explained by how much the edit changed the molecule's size, the action, and the round, none of which describes the target. That is not because the target is irrelevant: naming it outright separates them several times more sharply. For a pocket absent from the base, transfer must rely on a description rather than its identity. Among the exact-match descriptions we evaluate, the more specific ones rarely recur, whereas the coarser ones provide little separation; the signals that do transfer primarily describe the edit and the molecule it produced rather than the target. Under the representations and lookup scheme studied here, an experience base may be more useful for prioritizing an agent's own proposals for docking than for directly prescribing what to build.
ER -