Research map / Probabilistic and scientific modeling / Molecules and proteins
FABind: Fast and Accurate Protein-Ligand Binding
Qizhi Pei, Kaiyuan Gao, Lijun Wu, Jinhua Zhu, Yingce Xia, Shufang Xie, Tao Qin, Kun He, Tie-Yan Liu, Rui Yan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use ligand information to guide pocket prediction (ablation) | FABind pocket prediction · Pocket prediction | DCC % below 3Å: 36.9 -> 42.7 | Improved |
| Add the pocket center constraint (ablation) | FABind pocket prediction · Pocket prediction | DCC % below 3Å: 8.8 -> 42.7 | Improved |
| Add scheduled sampling to docking training (ablation) | FABind · PDBbind v2020 test | Ligand RMSD % below 2Å: 28.7 -> 33.1 | Improved |
| Use the full distance-map loss rather than coordinate loss alone (ablation) | FABind · PDBbind v2020 test | Ligand RMSD mean: 6.9 -> 6.4 | Improved |
| Add the remaining distance-map loss terms beyond the first term (ablation) | FABind · PDBbind v2020 test | Ligand RMSD mean: 6.5 -> 6.4 | Improved |
| Add cross-attention update (ablation) | FABind · PDBbind v2020 test | Ligand RMSD % below 2Å: 21.4 -> 33.1 | Improved |
| Add independent message passing (ablation) | FABind · PDBbind v2020 test | Ligand RMSD % below 2Å: 22.6 -> 33.1 | Improved |
| Add iterative refinement (ablation) | FABind · PDBbind v2020 test | Ligand RMSD % below 2Å: 22.5 -> 33.1 | Improved |
| Use ESM-2 protein features (ablation) | FABind · PDBbind v2020 test | Ligand RMSD 75th percentile: 8.0 -> 6.7 | Improved |
| Add post-optimization (ablation) | FABind · PDBbind v2020 test | Ligand RMSD mean: 6.4 -> 6.6 | Hurt |
| Increase iterative refinement from 1 to 8 rounds (ablation) | FABind · PDBbind v2020 test, inference | Ligand RMSD mean: 7.3 -> 6.4 | Improved |
| Use 8 refinement rounds instead of 4 (ablation) | FABind · PDBbind v2020 test, inference | Ligand RMSD mean: 6.5 -> 6.4 | No clear effect |
Later papers that compared with this method
2 later accepted papers report FABind+ as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Fast and Accurate Blind Flexible Docking (ICLR 2025 · better on 2 of 2 shared results)
- DeltaDock: A Unified Framework for Accurate, Efficient, and Physically Reliable Molecular Docking (NeurIPS 2024 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DiffDock (10) — DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking (ICLR 2023) · better on 2 of 3 shared results
- E3Bind — E3Bind: An End-to-End Equivariant Network for Protein-Ligand Docking (ICLR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Fast and Accurate Blind Flexible Docking (ICLR 2025 · similarity 0.91)
- DeltaDock: A Unified Framework for Accurate, Efficient, and Physically Reliable Molecular Docking (NeurIPS 2024 · similarity 0.89)
- Group Ligands Docking to Protein Pockets (ICLR 2025 · similarity 0.88)
- UniSite: The First Cross-Structure Dataset and Learning Framework for End-to-End Ligand Binding Site Detection (NeurIPS 2025 · similarity 0.86)
- Protein-ligand binding representation learning from fine-grained interactions (ICLR 2024 · similarity 0.86)
- Towards precision protein-ligand affinity prediction benchmark: A Complete and Modification-Aware DAVIS Dataset (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
