Research map / Graph learning / Graph self-supervised learning
DrugOOD: Out-of-Distribution Dataset Curator and Benchmark for AI-Aided Drug Discovery – a Focus on Affinity Prediction Problems with Noise Annotations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Mixup augmentation instead of ERM | GIN on DrugOOD LBAP IC50 assay, scaffold, and size splits · DrugOOD LBAP IC50 OOD | AUROC: 71.98 (0.29) -> 71.34 (0.41); 69.54 (0.52) -> 69.87 (0.17); 67.48 (0.47) -> 67.73 (0.27) | No clear effect |
| Use IRM instead of ERM | GIN on DrugOOD LBAP IC50 · Assay, scaffold, and size OOD splits | AUROC: 71.98 (0.29) -> 69.22 (0.51); 67.48 (0.47) -> 57.00 (0.39); 68.74 (0.37) -> 59.99 (0.48) | Hurt |
| Use DeepCoral instead of ERM | GIN on DrugOOD LBAP IC50 · General-noise assay, scaffold, and size OOD splits | AUROC: 67.48 (0.47) -> 57.31 (0.44); 66.01 (0.14) -> 61.09 (0.23); 68.74 (0.37) -> 62.01 (0.38) | Hurt |
| Use DANN instead of ERM | GIN on DrugOOD LBAP IC50 · Core assay and scaffold OOD splits | AUROC: 71.98 (0.29) -> 70.08 (0.65); 69.88 (0.13) -> 65.12 (0.94) | Hurt |
| Use GroupDRO instead of ERM | GIN on DrugOOD LBAP IC50 · General-noise assay, scaffold, and size OOD splits | AUROC: 67.48 (0.47) -> 60.90 (1.23); 66.01 (0.14) -> 61.41 (0.10); 68.74 (0.37) -> 62.15 (0.54) | Hurt |
| Evaluate on OOD rather than in-distribution data | ERM on DrugOOD LBAP and SBAP · IC50 assay, size, and protein domains | AUROC: 89.62 (2.04) -> 71.98 (0.29); 92.35 (0.15) -> 67.48 (0.47); 90.71 (0.29) -> 68.87 (0.53) | Hurt |
Closest papers by meaning
- BOOM: Benchmarking Out-Of-distribution Molecular Property Predictions of Machine Learning Models (NeurIPS 2025 · similarity 0.88)
- Knowledge Enhanced Representation Learning for Drug Discovery (AAAI 2024 · similarity 0.86)
- PharmacoMatch: Efficient 3D Pharmacophore Screening via Neural Subgraph Matching (ICLR 2025 · similarity 0.85)
- M²N: A Progressive Macro-to-Micro 3D Modeling Scheme for Unveiling Drug-Target Affinity (AAAI 2025 · similarity 0.85)
- BounDr.E: Predicting Drug-likeness via Biomedical Knowledge Alignment and EM-like One-Class Boundary Optimization (ICML 2025 · similarity 0.85)
- WelQrate: Defining the Gold Standard in Small Molecule Drug Discovery Benchmarking (NeurIPS 2024 · similarity 0.84)
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.