KinetiDiff: Docking-Guided Diffusion for De Novo ACVR1 Inhibitor Design in Fibrodysplasia Ossificans Progressiva
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915950579679232 |
|---|---|
| author | Patel, Aaryan |
| author_facet | Patel, Aaryan |
| contents | We present KinetiDiff, a structure-based framework for de novo kinase inhibitor design that integrates a Geometry-Complete Diffusion Model with real-time AutoDock Vina gradient guidance. By injecting physics-based docking gradients into the diffusion denoising loop, KinetiDiff steers molecule generation toward high-affinity conformations for ACVR1 (ALK2), the causative kinase in Fibrodysplasia Ossificans Progressiva. From 10,000 diffusion samples, the framework produced 9,997 valid molecules. The best candidate achieved $-11.05$ kcal/mol (pKd = 8.10), a 19.2% improvement over the crystallographic reference. The top 100 candidates all exceed the reference, with 100% Lipinski compliance, median synthetic accessibility of 2.67, and internal diversity of 0.790. Systematic ablation across four guidance strategies--Vina-Direct (physics), HNN-Denovo (neural proxy), multi-objective, and unguided--demonstrates that real-time docking guidance dominates on all metrics. We evaluate HNN-Denovo as a computationally efficient alternative (60-fold speedup per step), revealing a domain-mismatch limitation (r = 0.224 correlation with Vina) that explains its inferior performance. These results establish gradient-guided geometric diffusion as a practical approach for generating potent, synthetically accessible inhibitors against rare-disease kinase targets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20886 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | KinetiDiff: Docking-Guided Diffusion for De Novo ACVR1 Inhibitor Design in Fibrodysplasia Ossificans Progressiva Patel, Aaryan Chemical Physics Machine Learning I.2.6; I.5.1; I.5.4; J.3 We present KinetiDiff, a structure-based framework for de novo kinase inhibitor design that integrates a Geometry-Complete Diffusion Model with real-time AutoDock Vina gradient guidance. By injecting physics-based docking gradients into the diffusion denoising loop, KinetiDiff steers molecule generation toward high-affinity conformations for ACVR1 (ALK2), the causative kinase in Fibrodysplasia Ossificans Progressiva. From 10,000 diffusion samples, the framework produced 9,997 valid molecules. The best candidate achieved $-11.05$ kcal/mol (pKd = 8.10), a 19.2% improvement over the crystallographic reference. The top 100 candidates all exceed the reference, with 100% Lipinski compliance, median synthetic accessibility of 2.67, and internal diversity of 0.790. Systematic ablation across four guidance strategies--Vina-Direct (physics), HNN-Denovo (neural proxy), multi-objective, and unguided--demonstrates that real-time docking guidance dominates on all metrics. We evaluate HNN-Denovo as a computationally efficient alternative (60-fold speedup per step), revealing a domain-mismatch limitation (r = 0.224 correlation with Vina) that explains its inferior performance. These results establish gradient-guided geometric diffusion as a practical approach for generating potent, synthetically accessible inhibitors against rare-disease kinase targets. |
| title | KinetiDiff: Docking-Guided Diffusion for De Novo ACVR1 Inhibitor Design in Fibrodysplasia Ossificans Progressiva |
| topic | Chemical Physics Machine Learning I.2.6; I.5.1; I.5.4; J.3 |
| url | https://arxiv.org/abs/2604.20886 |