Equivariant Scalar Fields for Molecular Docking with Fast Fourier Transforms
This work addresses throughput limitations in virtual screening for drug discovery, representing an incremental improvement with a novel method for a known bottleneck.
The authors tackled the slow optimization of scoring functions in molecular docking by learning a scoring function with a form that enables rapid optimization using fast Fourier transforms, achieving similar performance to Vina and Gnina but faster on crystal structures and more robust on predicted structures.
Molecular docking is critical to structure-based virtual screening, yet the throughput of such workflows is limited by the expensive optimization of scoring functions involved in most docking algorithms. We explore how machine learning can accelerate this process by learning a scoring function with a functional form that allows for more rapid optimization. Specifically, we define the scoring function to be the cross-correlation of multi-channel ligand and protein scalar fields parameterized by equivariant graph neural networks, enabling rapid optimization over rigid-body degrees of freedom with fast Fourier transforms. The runtime of our approach can be amortized at several levels of abstraction, and is particularly favorable for virtual screening settings with a common binding pocket. We benchmark our scoring functions on two simplified docking-related tasks: decoy pose scoring and rigid conformer docking. Our method attains similar but faster performance on crystal structures compared to the widely-used Vina and Gnina scoring functions, and is more robust on computationally predicted structures. Code is available at https://github.com/bjing2016/scalar-fields.