Ophiuchus: Scalable Modeling of Protein Structures through Hierarchical Coarse-graining SO(3)-Equivariant Autoencoders

TL;DR

Ophiuchus is a scalable, hierarchical, and equivariant autoencoder for protein structures that captures high-level patterns and enables efficient modeling, interpolation, and generation of proteins.

Contribution

It introduces a novel SO(3)-equivariant coarse-graining autoencoder that operates on all-atom protein structures, addressing limitations of traditional graph-based models.

Findings

Effective reconstruction across various compression rates.

Latent space enables conformational interpolation.

Diffusion models in latent space facilitate protein generation.

Abstract

Three-dimensional native states of natural proteins display recurring and hierarchical patterns. Yet, traditional graph-based modeling of protein structures is often limited to operate within a single fine-grained resolution, and lacks hourglass neural architectures to learn those high-level building blocks. We narrow this gap by introducing Ophiuchus, an SO(3)-equivariant coarse-graining model that efficiently operates on all-atom protein structures. Our model departs from current approaches that employ graph modeling, instead focusing on local convolutional coarsening to model sequence-motif interactions with efficient time complexity in protein length. We measure the reconstruction capabilities of Ophiuchus across different compression rates, and compare it to existing models. We examine the learned latent space and demonstrate its utility through conformational interpolation. Finally, we leverage denoising diffusion probabilistic models (DDPM) in the latent space to efficiently sample protein structures. Our experiments demonstrate Ophiuchus to be a scalable basis for efficient protein modeling and generation.