Design of metalloproteins and novel protein folds using variational autoencoders

TL;DR

This paper introduces a deep learning approach using variational autoencoders to generate novel protein sequences with desired properties, enabling scalable and automated protein design.

Contribution

It presents a novel application of variational autoencoders for protein design, including adding metal binding sites and creating new protein topologies, advancing automated protein engineering.

Findings

Successfully added metal binding sites without human intervention

Generated stable novel protein structures via a learned grammar

Model scales well and confines the search space effectively

Abstract

The design of novel proteins has many applications but remains an attritional process with success in isolated cases. Meanwhile, deep learning technologies have exploded in popularity in recent years and are increasingly applicable to biology due to the rise in available data. We attempt to link protein design and deep learning by using variational autoencoders to generate protein sequences conditioned on desired properties. Potential copper and calcium binding sites are added to non-metal binding proteins without human intervention and compared to a hidden Markov model. In another use case, a grammar of protein structures is developed and used to produce sequences for a novel protein topology. One candidate structure is found to be stable by molecular dynamics simulation. The ability of our model to confine the vast search space of protein sequences and to scale easily has the potential to assist in a variety of protein design tasks.