AI Insight
Researchers developed Harmony, a new computational method for predicting how drug molecules bind to proteins by accounting for both the ligand's position and the protein's flexibility. The method uses harmonic torsional diffusion that respects the periodic geometry of molecular rotations, parameterizing angular variables on a circle rather than treating them as linear coordinates. Testing on the PDBBind benchmark showed Harmony improves accuracy in predicting ligand binding poses and reconstructing protein pocket structures compared to existing flexible docking methods, while also generating more physically valid molecular complexes on the PoseBusters dataset.
Why it matters
This advance could accelerate drug discovery by more accurately predicting how potential drug molecules interact with flexible protein targets. Better computational docking methods reduce the need for expensive experimental screening and help identify promising drug candidates earlier in the development process.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
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, whose noise-level dependence is supplied analytically by the heat semigroup 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.
Source: Harmonic Torsional Diffusion for Protein-Ligand Flexible Docking