Beyond Affinity: A Benchmark of 1D, 2D, and 3D Methods Reveals Critical Trade-offs in Structure-Based Drug Design

TL;DR

This paper benchmarks 15 structure-based drug design models across 1D, 2D, and 3D methods, revealing their strengths and weaknesses in binding affinity, chemical validity, and pose quality to guide future model development.

Contribution

It provides the first comprehensive cross-algorithm benchmark for SBDD models, comparing their performance on pharmaceutical properties and docking metrics.

Findings

3D models excel in binding affinity but lack in chemical validity.

1D models are reliable in molecular metrics but less optimal in binding.

2D models balance chemical validity with moderate binding scores.

Abstract

Currently, the field of structure-based drug design is dominated by three main types of algorithms: search-based algorithms, deep generative models, and reinforcement learning. While existing works have typically focused on comparing models within a single algorithmic category, cross-algorithm comparisons remain scarce. In this paper, to fill the gap, we establish a benchmark to evaluate the performance of fifteen models across these different algorithmic foundations by assessing the pharmaceutical properties of the generated molecules and their docking affinities and poses with specified target proteins. We highlight the unique advantages of each algorithmic approach and offer recommendations for the design of future SBDD models. We emphasize that 1D/2D ligand-centric drug design methods can be used in SBDD by treating the docking function as a black-box oracle, which is typically neglected. Our evaluation reveals distinct patterns across model categories. 3D structure-based models excel in binding affinities but show inconsistencies in chemical validity and pose quality. 1D models demonstrate reliable performance in standard molecular metrics but rarely achieve optimal binding affinities. 2D models offer balanced performance, maintaining high chemical validity while achieving moderate binding scores. Through detailed analysis across multiple protein targets, we identify key improvement areas for each model category, providing insights for researchers to combine strengths of different approaches while addressing their limitations. All the code that are used for benchmarking is available in https://github.com/zkysfls/2025-sbdd-benchmark