Generating tertiary protein structures via interpretable graph variational autoencoders

Xiaojie Guo; Yuanqi Du; Sivani Tadepalli; Liang Zhao; Amarda Shehu

doi:10.1093/bioadv/vbab036

Generating tertiary protein structures via interpretable graph variational autoencoders

Bioinform Adv. 2021 Nov 29;1(1):vbab036. doi: 10.1093/bioadv/vbab036. eCollection 2021.

Authors

Xiaojie Guo¹, Yuanqi Du², Sivani Tadepalli², Liang Zhao³, Amarda Shehu⁴

Affiliations

¹ Department of Information Sciences and Technology, George Mason University, Fairfax, VA 22030, USA.
² Department of Computer Science, George Mason University, Fairfax, VA 22030, USA.
³ Department of Computer Science, Emory University, Atlanta, GA 30322, USA.
⁴ Department of Bioengineering, George Mason University, Fairfax, VA 22030, USA.

Abstract

Motivation: Modeling the structural plasticity of protein molecules remains challenging. Most research has focused on obtaining one biologically active structure. This includes the recent AlphaFold2 that has been hailed as a breakthrough for protein modeling. Computing one structure does not suffice to understand how proteins modulate their interactions and even evade our immune system. Revealing the structure space available to a protein remains challenging. Data-driven approaches that learn to generate tertiary structures are increasingly garnering attention. These approaches exploit the ability to represent tertiary structures as contact or distance maps and make direct analogies with images to harness convolution-based generative adversarial frameworks from computer vision. Since such opportunistic analogies do not allow capturing highly structured data, current deep models struggle to generate physically realistic tertiary structures.

Results: We present novel deep generative models that build upon the graph variational autoencoder framework. In contrast to existing literature, we represent tertiary structures as 'contact' graphs, which allow us to leverage graph-generative deep learning. Our models are able to capture rich, local and distal constraints and additionally compute disentangled latent representations that reveal the impact of individual latent factors. This elucidates what the factors control and makes our models more interpretable. Rigorous comparative evaluation along various metrics shows that the models, we propose advance the state-of-the-art. While there is still much ground to cover, the work presented here is an important first step, and graph-generative frameworks promise to get us to our goal of unraveling the exquisite structural complexity of protein molecules.

Availability and implementation: Code is available at https://github.com/anonymous1025/CO-VAE.

Supplementary information: Supplementary data are available at Bioinformatics Advances online.