PDB-Struct: A Comprehensive Benchmark for Structure-based Protein Design

TL;DR

This paper introduces PDB-Struct, a comprehensive benchmark with novel metrics for evaluating structure-based protein design methods, addressing limitations of existing validation approaches and enabling fair comparison of recent algorithms.

Contribution

The paper presents a new benchmark dataset and two innovative metrics for assessing protein design methods, improving evaluation accuracy beyond traditional recovery and perplexity measures.

Findings

ByProt, ProteinMPNN, and ESM-IF perform well on the benchmark.

ESM-Design and AF-Design underperform on refoldability.

Some methods show high sequence recovery but low true foldability.

Abstract

Structure-based protein design has attracted increasing interest, with numerous methods being introduced in recent years. However, a universally accepted method for evaluation has not been established, since the wet-lab validation can be overly time-consuming for the development of new algorithms, and the $\textit{in silico}$ validation with recovery and perplexity metrics is efficient but may not precisely reflect true foldability. To address this gap, we introduce two novel metrics: refoldability-based metric, which leverages high-accuracy protein structure prediction models as a proxy for wet lab experiments, and stability-based metric, which assesses whether models can assign high likelihoods to experimentally stable proteins. We curate datasets from high-quality CATH protein data, high-throughput $\textit{de novo}$ designed proteins, and mega-scale experimental mutagenesis experiments, and in doing so, present the $\textbf{PDB-Struct}$ benchmark that evaluates both recent and previously uncompared protein design methods. Experimental results indicate that ByProt, ProteinMPNN, and ESM-IF perform exceptionally well on our benchmark, while ESM-Design and AF-Design fall short on the refoldability metric. We also show that while some methods exhibit high sequence recovery, they do not perform as well on our new benchmark. Our proposed benchmark paves the way for a fair and comprehensive evaluation of protein design methods in the future. Code is available at https://github.com/WANG-CR/PDB-Struct.