RangeFinder: Predicting Ligand Property Ranges from Protein Pockets
Abstract
Identifying ligands for a protein target requires searching a vast chemical space for a comparatively small set of compatible molecules. Existing machine learning approaches primarily address this problem by generating candidate ligands or scoring individual protein-ligand pairs, but provide limited information about the broader chemical space compatible with a binding pocket. We introduce RangeFinder, an SE(3)-equivariant transformer pocket-level model, trained and evaluated on BioLiP2 data, that predicts ranges over nine physicochemical ligand properties directly from pocket structure, chemistry, and evolutionary context. Using an intersection-over-union objective and evaluation metric (range-IoU), RangeFinder achieves a range-IoU of 0.464 ± 0.002, outperforming all deployable baselines while requiring only $12.2$ ms per pocket. We also quantify our method's ability to reduce the ligand search space while retaining compatible binders; in this regard, RangeFinder achieves a Reduction-Recall F-Score of 0.543 ± 0.011 --- once again outperforming all deployable baselines. Ablation experiments indicate that RangeFinder's performance depends more strongly on the density and quality of ligand supervision than on any individual input modality or specialized training strategy. Together, these results position RangeFinder as an inexpensive and effective molecular filtering prior, with demonstrated potential in narrowing the ligand search space of target pockets before docking, generation, or synthesis. The code for this project can be found at: https://github.com/alannadels/RangeFinder.
Related
- What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design
- Dynamic Protein Adduct Formation by Methylmercury: Molecular Targets, Ligand Exchange, and Toxicological Consequences
- Transition Intermediates Encode Ligand Efficacy at the μ-Opioid Receptor-Gi Protein Complex
- A Proximity-Driven Functional Screening Platform for the Discovery of Noninhibitory USP7 Ligand Enabling Targeted Protein Stabilization
- Frustration Quenching and Network Topology of the Energy Landscape as Primary Determinants of Protein-Ligand Binding Pose Prediction by Deep Learning Models
BibTeX
@misc{ruvalcaba2026rangefinder,
title = {{RangeFinder: Predicting Ligand Property Ranges from Protein Pockets}},
author = {Alan Nadelsticher Ruvalcaba and Chris Alvin and Supratik Mukhopadhyay and Adam Bess},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.03379},
eprint = {2611.03379},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.03379},
}
RIS
TY - UNPB
TI - RangeFinder: Predicting Ligand Property Ranges from Protein Pockets
AU - Alan Nadelsticher Ruvalcaba
AU - Chris Alvin
AU - Supratik Mukhopadhyay
AU - Adam Bess
PY - 2026
DA - 2026/11/02
PB - arXiv
DO - 10.48550/arxiv.2611.03379
UR - https://arxiv.org/abs/2611.03379
AB - Identifying ligands for a protein target requires searching a vast chemical space for a comparatively small set of compatible molecules. Existing machine learning approaches primarily address this problem by generating candidate ligands or scoring individual protein-ligand pairs, but provide limited information about the broader chemical space compatible with a binding pocket. We introduce RangeFinder, an SE(3)-equivariant transformer pocket-level model, trained and evaluated on BioLiP2 data, that predicts ranges over nine physicochemical ligand properties directly from pocket structure, chemistry, and evolutionary context. Using an intersection-over-union objective and evaluation metric (range-IoU), RangeFinder achieves a range-IoU of 0.464 ± 0.002, outperforming all deployable baselines while requiring only $12.2$ ms per pocket. We also quantify our method's ability to reduce the ligand search space while retaining compatible binders; in this regard, RangeFinder achieves a Reduction-Recall F-Score of 0.543 ± 0.011 --- once again outperforming all deployable baselines. Ablation experiments indicate that RangeFinder's performance depends more strongly on the density and quality of ligand supervision than on any individual input modality or specialized training strategy. Together, these results position RangeFinder as an inexpensive and effective molecular filtering prior, with demonstrated potential in narrowing the ligand search space of target pockets before docking, generation, or synthesis. The code for this project can be found at: https://github.com/alannadels/RangeFinder.
ER -