# PoseX: AI Defeats Physics Approaches on Protein-Ligand Cross Docking

arXiv: https://arxiv.org/abs/2505.01700 · HTML: https://mlsanity.com/paper/2505.01700/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 45 references that are in the map.
- First posted on arXiv 3 May 2025.

## Builds on (most cited first, 10 of 10)

- [DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking](https://mlsanity.com/paper/2210.01776/) (2022)
- [EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction](https://mlsanity.com/paper/2202.05146/) (2022)
- [MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design](https://mlsanity.com/paper/2203.14500/) (2022)
- [Reinforced Genetic Algorithm for Structure-based Drug Design](https://mlsanity.com/paper/2211.16508/) (2022)
- [Molformer: Motif-based Transformer on 3D Heterogeneous Molecular Graphs](https://mlsanity.com/paper/2110.01191/) (2021)
- [State-specific protein-ligand complex structure prediction with a multi-scale deep generative model](https://mlsanity.com/paper/2209.15171/) (2022)
- [Deep Confident Steps to New Pockets: Strategies for Docking Generalization](https://mlsanity.com/paper/2402.18396/) (2024)
- [FABind: Fast and Accurate Protein-Ligand Binding](https://mlsanity.com/paper/2310.06763/) (2023)
- [Uni-Mol Docking V2: Towards Realistic and Accurate Binding Pose Prediction](https://mlsanity.com/paper/2405.11769/) (2024)
- [Instructor-inspired Machine Learning for Robust Molecular Property Prediction](https://mlsanity.com/paper/2304.03906/) (2023)

## Built on by (most cited first, 7 of 7)

- [Proteo-R1: Reasoning Foundation Models for De Novo Protein Design](https://mlsanity.com/paper/2605.02937/) (2026)
- [SigmaDock: Untwisting Molecular Docking With Fragment-Based SE(3) Diffusion](https://mlsanity.com/paper/2511.04854/) (2025)
- [MC-GNNAS-Dock: Multi-criteria GNN-based Algorithm Selection for Molecular Docking](https://mlsanity.com/paper/2509.26377/) (2025)
- [CovDocker: Benchmarking Covalent Drug Design with Tasks, Datasets, and Solutions](https://mlsanity.com/paper/2506.21085/) (2025)
- [Can AI-predicted complexes teach machine learning to compute drug binding affinity?](https://mlsanity.com/paper/2507.07882/) (2025)
- [Deciphering Fingerprints of 3D Molecular Surfaces for Accurate Epitope Prediction](https://mlsanity.com/paper/2606.23830/) (2026)
- [Accurate structural modeling of chemically diverse molecular interfaces with Vilya-2](https://mlsanity.com/paper/2607.25156/) (2026)

## Abstract (arXiv)

Existing protein-ligand docking studies typically focus on the self-docking scenario, which is less practical in real applications. Moreover, some studies involve heavy frameworks requiring extensive training, posing challenges for convenient and efficient assessment of docking methods. To fill these gaps, we design PoseX, an open-source benchmark to evaluate both self-docking and cross-docking, enabling a practical and comprehensive assessment of algorithmic advances. Specifically, we curated a novel dataset comprising 718 entries for self-docking and 1,312 entries for cross-docking; second, we incorporated 23 docking methods in three methodological categories, including physics-based methods (e.g., Schr"odinger Glide), AI docking methods (e.g., DiffDock) and AI co-folding methods (e.g., AlphaFold3); third, we developed a relaxation method for post-processing to minimize conformational energy and refine binding poses; fourth, we built a leaderboard to rank submitted models in real-time. We derived some key insights and conclusions from extensive experiments: (1) AI approaches have consistently outperformed physics-based methods in overall docking success rate. (2) Most intra- and intermolecular clashes of AI approaches can be greatly alleviated with relaxation, which means combining AI modeling with physics-based post-processing could achieve excellent performance. (3) AI co-folding methods exhibit ligand chirality issues, except for Boltz-1x, which introduced physics-inspired potentials to fix hallucinations, suggesting modeling on stereochemistry improves the structural plausibility markedly. (4) Specifying binding pockets significantly promotes docking performance, indicating that pocket information can be leveraged adequately, particularly for AI co-folding methods, in future modeling efforts. The code, dataset, and leaderboard are released at https://github.com/CataAI/PoseX.