H-Packer: Holographic Rotationally Equivariant Convolutional Neural Network for Protein Side-Chain Packing
This work addresses protein structure modeling for computational biology, offering an incremental improvement over existing deep learning methods.
The authors tackled protein side-chain packing by framing it as a joint regression over dihedral angles, proposing H-Packer, a two-stage algorithm using rotationally equivariant neural networks. The method showed favorable performance against physics-based algorithms and was competitive with other deep learning solutions on CASP13 and CASP14 targets.
Accurately modeling protein 3D structure is essential for the design of functional proteins. An important sub-task of structure modeling is protein side-chain packing: predicting the conformation of side-chains (rotamers) given the protein's backbone structure and amino-acid sequence. Conventional approaches for this task rely on expensive sampling procedures over hand-crafted energy functions and rotamer libraries. Recently, several deep learning methods have been developed to tackle the problem in a data-driven way, albeit with vastly different formulations (from image-to-image translation to directly predicting atomic coordinates). Here, we frame the problem as a joint regression over the side-chains' true degrees of freedom: the dihedral $χ$ angles. We carefully study possible objective functions for this task, while accounting for the underlying symmetries of the task. We propose Holographic Packer (H-Packer), a novel two-stage algorithm for side-chain packing built on top of two light-weight rotationally equivariant neural networks. We evaluate our method on CASP13 and CASP14 targets. H-Packer is computationally efficient and shows favorable performance against conventional physics-based algorithms and is competitive against alternative deep learning solutions.