Harmonic Torsional Diffusion for Protein-Ligand Flexible Docking
Abstract
Molecular docking requires reasoning jointly about ligand pose and protein flexibility. Most diffusion-based docking models predict torsional updates with generic Euclidean heads that ignore the periodic geometry of angular variables. This mismatch is especially limiting in flexible docking, where ligand conformations and pocket side chains co-adapt to form the bound complex. Here, we introduce Harmony, a harmonic torsional diffusion framework for flexible protein-ligand docking. Harmony parameterizes ligand and side-chain torsional score fields as derivatives of learned harmonic potentials on the circle, which are noised mode by mode in the Fourier basis owing to the properties of variance-exploding diffusion on the torus. This construction makes periodicity explicit and gives the model a frequency-aware inductive bias over rotameric motion. On the PDBBind benchmark, Harmony improves ligand pose accuracy and pocket all-atom reconstruction over recent flexible docking methods. On PoseBusters, it improves the physical validity of generated complexes. Case studies on EBNA1 and KRAS G12D illustrate the method's behavior on a polar and a shallow binding site, respectively. Together, these results indicate that aligning the score parameterization with the geometry of the diffusion process is a simple and effective lever for improving flexible docking.
Related
- Geometric Relevance Is Not Sufficient: Evaluating Focused Post-Training Supervision in Diffusion Docking
- Unified Sampling and Ranking for Protein Docking With DFMDock
- Structural insights into an aptainhibitor of the SARS-CoV-2 E protein using molecular docking, molecular dynamics, and biophysical analyses
- Improving scoring functions for protein-protein docking with LambdaLoss
- De novo design of flexible protein interactions with GuideFlip
BibTeX
@misc{zhdanov2026harmonic,
title = {{Harmonic Torsional Diffusion for Protein-Ligand Flexible Docking}},
author = {Maksim Zhdanov and Pavel Strashnov and Vladislav Kurenkov},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2608.20366},
eprint = {2608.20366},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.20366},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Harmonic Torsional Diffusion for Protein-Ligand Flexible Docking
AU - Maksim Zhdanov
AU - Pavel Strashnov
AU - Vladislav Kurenkov
PY - 2026
DA - 2026/10/18
PB - arXiv
DO - 10.48550/arxiv.2608.20366
UR - https://arxiv.org/abs/2608.20366
AB - Molecular docking requires reasoning jointly about ligand pose and protein flexibility. Most diffusion-based docking models predict torsional updates with generic Euclidean heads that ignore the periodic geometry of angular variables. This mismatch is especially limiting in flexible docking, where ligand conformations and pocket side chains co-adapt to form the bound complex. Here, we introduce Harmony, a harmonic torsional diffusion framework for flexible protein-ligand docking. Harmony parameterizes ligand and side-chain torsional score fields as derivatives of learned harmonic potentials on the circle, which are noised mode by mode in the Fourier basis owing to the properties of variance-exploding diffusion on the torus. This construction makes periodicity explicit and gives the model a frequency-aware inductive bias over rotameric motion. On the PDBBind benchmark, Harmony improves ligand pose accuracy and pocket all-atom reconstruction over recent flexible docking methods. On PoseBusters, it improves the physical validity of generated complexes. Case studies on EBNA1 and KRAS G12D illustrate the method's behavior on a polar and a shallow binding site, respectively. Together, these results indicate that aligning the score parameterization with the geometry of the diffusion process is a simple and effective lever for improving flexible docking.
ER -