Importance Weighted Expectation-Maximization for Protein Sequence Design

TL;DR

This paper introduces IsEM-Pro, a novel latent generative model combined with Markov random fields, to efficiently generate diverse, high-fitness protein sequences, outperforming previous methods significantly.

Contribution

The paper presents IsEM-Pro, a new approach that integrates Monte Carlo EM with MRFs for protein sequence design, enhancing diversity and fitness in generated sequences.

Findings

Outperforms previous methods by at least 55% on average fitness score

Generates more diverse and novel protein sequences

Effective in eight different protein design tasks

Abstract

Designing protein sequences with desired biological function is crucial in biology and chemistry. Recent machine learning methods use a surrogate sequence-function model to replace the expensive wet-lab validation. How can we efficiently generate diverse and novel protein sequences with high fitness? In this paper, we propose IsEM-Pro, an approach to generate protein sequences towards a given fitness criterion. At its core, IsEM-Pro is a latent generative model, augmented by combinatorial structure features from a separately learned Markov random fields (MRFs). We develop an Monte Carlo Expectation-Maximization method (MCEM) to learn the model. During inference, sampling from its latent space enhances diversity while its MRFs features guide the exploration in high fitness regions. Experiments on eight protein sequence design tasks show that our IsEM-Pro outperforms the previous best methods by at least 55% on average fitness score and generates more diverse and novel protein sequences.