Unified rational protein engineering with sequence-only deep representation learning

Abstract

AbstractRational protein engineering requires a holistic understanding of protein function. Here, we apply deep learning to unlabelled amino acid sequences to distill the fundamental features of a protein into a statisticalrepresentationthat is semantically rich and structurally, evolutionarily, and biophysically grounded. We show that the simplest models built on top of thisunifiedrepresentation (UniRep) are broadly applicable and generalize to unseen regions of sequence space. Our data-driven approach reaches near state-of-the-art or superior performance predicting stability of natural andde novodesigned proteins as well as quantitative function of molecularly diverse mutants. UniRep further enables two orders of magnitude cost savings in a protein engineering task. We conclude UniRep is a versatile protein summary that can be applied across protein engineering informatics.