G-VAE, a Geometric Convolutional VAE for ProteinStructure Generation

TL;DR

This paper introduces G-VAE, a geometric convolutional variational auto-encoder that models protein structures as 3D curves, enabling efficient comparison, deformation, and generation of novel protein conformations.

Contribution

The work presents a novel geometric neural network framework combining SRVF representation and ResNets with a G-VAE that maps protein structures to a low-dimensional spherical latent space.

Findings

Successfully generates plausible new protein structures

Accurately predicts completions of corrupted structures

Handles large deformations efficiently

Abstract

Analyzing the structure of proteins is a key part of understanding their functions and thus their role in biology at the molecular level. In addition, design new proteins in a methodical way is a major engineering challenge. In this work, we introduce a joint geometric-neural networks approach for comparing, deforming and generating 3D protein structures. Viewing protein structures as 3D open curves, we adopt the Square Root Velocity Function (SRVF) representation and leverage its suitable geometric properties along with Deep Residual Networks (ResNets) for a joint registration and comparison. Our ResNets handle better large protein deformations while being more computationally efficient. On top of the mathematical framework, we further design a Geometric Variational Auto-Encoder (G-VAE), that once trained, maps original, previously unseen structures, into a low-dimensional (latent) hyper-sphere. Motivated by the spherical structure of the pre-shape space, we naturally adopt the von Mises-Fisher (vMF) distribution to model our hidden variables. We test the effectiveness of our models by generating novel protein structures and predicting completions of corrupted protein structures. Experimental results show that our method is able to generate plausible structures, different from the structures in the training data.