Research map / Probabilistic and scientific modeling / Molecules and proteins
Learning Protein–Ligand Binding in Hyperbolic Space
Jianhui Wang, Wenyu Zhu, Bowen Gao, Xin Hong, Ya-Qin Zhang, Wei-Ying Ma, Yanyan Lan
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 |
|---|---|---|---|
| embed in hyperbolic space with cone loss and regularizers instead of Euclidean contrastive+rank (ablation) | HypSeek · DUD-E | EF1%: 42.87 -> 51.44 | Improved |
| add hyperbolic terms (Lcone, Rang, Rhet) (ablation) | HypSeek · JACS | Pearson r: 0.6839 -> 0.7518 | Improved |
| add hyperbolic terms (Lcone, Rang, Rhet) (ablation) | HypSeek · DUD-E | EF1%: 49.14 -> 51.44 | Improved |
| add angular regularizer Rang (ablation) | HypSeek · DUD-E | EF1%: 50.52 -> 51.44 | Improved |
| add heterogeneity regularizer Rhet (ablation) | HypSeek · JACS | Pearson r: 0.7047 -> 0.7518 | Improved |
| add protein sequence tower (ablation) | HypSeek · DUD-E | EF1%: 47.70 -> 51.44 | Improved |
| add protein sequence tower (ablation) | HypSeek · JACS | Spearman rho: 0.7050 -> 0.7580 | Improved |
| replace LigUnity with HypSeek (hyperbolic) | virtual screening · DUD-E | EF1%: 42.63 -> 51.44 | Improved |
| replace LigUnity with HypSeek (hyperbolic) | virtual screening · LIT-PCBA | EF1%: 6.47 -> 6.81 | Improved |
| HypSeek vs LigUnity ensemble on Merck | affinity ranking · Merck | Spearman rho: 0.5554 -> 0.5447 | Hurt |
| HypSeek vs FEP+ on Merck | affinity ranking · Merck | Pearson r: 0.6960 -> 0.6120 | Hurt |
| hyperbolic vs Euclidean on structurally similar ligand pairs | HypSeek · JACS pairwise | accuracy | Improved |
Closest papers by meaning
- DecoyDB: A Dataset for Graph Contrastive Learning in Protein-Ligand Binding Affinity Prediction (NeurIPS 2025 · similarity 0.87)
- CausalBind: Causal Modeling and Learning for Protein-Molecule Virtual Screening (NeurIPS 2026 · similarity 0.86)
- Latent Retrieval Augmented Generation of Cross-Domain Protein Binders (NeurIPS 2025 · similarity 0.86)
- AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation (NeurIPS 2025 · similarity 0.85)
- M²N: A Progressive Macro-to-Micro 3D Modeling Scheme for Unveiling Drug-Target Affinity (AAAI 2025 · similarity 0.85)
- UniSite: The First Cross-Structure Dataset and Learning Framework for End-to-End Ligand Binding Site Detection (NeurIPS 2025 · similarity 0.85)
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.
