SurfGraphPro: Integrating Protein Language Models with Geometric Deep Learning On Coarse Protein Surfaces for Binding Site Prediction
- 1Biosciences Division, Los Alamos National Laboratory
Abstract
Accurate and fast prediction of protein binding sites remains essential for understanding molecular interactions and facilitating protein engineering. Here, we present SurfGraphPro, a geometric deep learning approach that bridges protein language models with surface-based structural representations for a rapid binding interface prediction. Our method operates on coarse-graphed triangulated protein surfaces, utilizing solvent-excluded surface meshes downsampled into amino acid residue centered patches. Unlike existing approaches that rely on extensive physicochemical feature engineering, we leverage evolutionary information directly from protein language model embeddings, eliminating the need for computationally expensive multiple sequence alignments and hand-crafted features. This integration achieves on average ∼18−28 × speedup for proteins of ∼100 to a few 1000s of amino acids over current state-of-the-art surface-based model while maintaining comparable accuracy on diverse binding interfaces, including challenging antibody-antigen complexes. To our knowledge, this represents the first approach to integrate protein language model embeddings with coarse geometric surface representations for binding site prediction, demonstrating that learned evolutionary features coupled with geometric transformers can replace traditional feature engineering without sacrificing performance.
Related
- Integrating Multisource Target Evidence with Machine-Learning Potency Prediction for Alzheimer's Disease Drug Discovery
- SIMORGH: Ensemble-Aware Geometric Deep Learning for Apo-State and Cryptic Ligand Binding Site Prediction
- DeepAVPred: A Deep Learning Framework Integrating RECM-DWT and BiGRU for Accurate Antiviral Peptide Prediction
- GeoEPred: A multimodal structure-aware geometric deep learning framework for Gram-negative bacterial secreted effector prediction with sequence semantics
- Beyond prediction accuracy by integrating GWAASB into wheat stability assessment across environments in Saudi Arabia
BibTeX
@misc{habib2026surfgraphpro,
title = {{SurfGraphPro: Integrating Protein Language Models with Geometric Deep Learning On Coarse Protein Surfaces for Binding Site Prediction}},
author = {A. Habib and L.-W. Hung and K. Gibson and M. Dix and P. S. G. Chain and B. Hu},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.17.751511},
url = {https://www.biorxiv.org/content/10.64898/2026.09.17.751511v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - SurfGraphPro: Integrating Protein Language Models with Geometric Deep Learning On Coarse Protein Surfaces for Binding Site Prediction
AU - A. Habib
AU - L.-W. Hung
AU - K. Gibson
AU - M. Dix
AU - P. S. G. Chain
AU - B. Hu
PY - 2026
DA - 2026/09/17
PB - bioRxiv
DO - 10.64898/2026.09.17.751511
UR - https://www.biorxiv.org/content/10.64898/2026.09.17.751511v1
AB - Accurate and fast prediction of protein binding sites remains essential for understanding molecular interactions and facilitating protein engineering. Here, we present SurfGraphPro, a geometric deep learning approach that bridges protein language models with surface-based structural representations for a rapid binding interface prediction. Our method operates on coarse-graphed triangulated protein surfaces, utilizing solvent-excluded surface meshes downsampled into amino acid residue centered patches. Unlike existing approaches that rely on extensive physicochemical feature engineering, we leverage evolutionary information directly from protein language model embeddings, eliminating the need for computationally expensive multiple sequence alignments and hand-crafted features. This integration achieves on average ∼18−28 × speedup for proteins of ∼100 to a few 1000s of amino acids over current state-of-the-art surface-based model while maintaining comparable accuracy on diverse binding interfaces, including challenging antibody-antigen complexes. To our knowledge, this represents the first approach to integrate protein language model embeddings with coarse geometric surface representations for binding site prediction, demonstrating that learned evolutionary features coupled with geometric transformers can replace traditional feature engineering without sacrificing performance.
ER -