Molecular Property Prediction on MoleculeNet BACE (scaffold)
89ROC-AUCMolCLR
Evaluation Results
| Method | Links | |
|---|---|---|
| MolCLR2023.04 | 89 | |
| ChemBFN2024.09 | 88.4 | |
| DumplingGNN2024.09 | 88.2 | |
| RF2023.08 | 86.7 | |
| MolBERT2023.04 | 86.6 | |
| SVM2023.08 | 86.2 | |
| Hu et al.2023.08 | 85.9 | |
| Hu et al.2023.04 | 85.9 | |
| MolXPT2024.09 | 85.7 | |
| GEM2023.04 | 85.6 | |
| Uni-Mol2024.09 | 85.6 | |
| KPGT2023.04 | 85.5 | |
| MPTr+CIPEBackbone=MPTr, Positional encoding=CIPE2026.06 | 85.5 | |
| D-MPNN2023.08 | 85.3 | |
| Hi-GMAE-FPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 85 | |
| GROVER_largeModel scale=large2024.09 | 84.5 | |
| GINGraph-level pre-training=Supervised, Node-level pre-training=Context Prediction2019.05 | 84.5 | |
| GraphLogPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 83.5 | |
| GraphLoG2026.06 | 83.5 | |
| SELFormer2023.04 | 83.2 | |
| GraphMAEPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 83.1 | |
| GraphMAE2026.06 | 83.1 | |
| GraSPNet2026.02 | 82.9 | |
| GraphLOG2026.02 | 82.8 | |
| Mole-BERT2026.02 | 82.8 | |
| Hi-GMAE-GPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 82.6 | |
| GEM2024.09 | 82.4 | |
| APTBase Model=Mole-BERT2023.11 | 82.3 | |
| G2PT2026.06 | 82.3 | |
| APTBase Model=EdgePred2023.11 | 81.8 | |
| GraphMAE2026.02 | 81.7 | |
| SimSGT2026.02 | 81.5 | |
| GraphMVP-C2023.04 | 81.2 | |
| MolCLR2024.09 | 81.2 | |
| GraphMVP2024.09 | 81 | |
| GraphMAE22026.02 | 81 | |
| GraphMAE2Pre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 81 | |
| D-MPNN2023.04 | 80.9 | |
| D-MPNN2024.09 | 80.9 | |
| Mole-BERT2023.11 | 80.8 | |
| Mole-BERTPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 80.8 | |
| GPSEaugmented2026.06 | 80.8 | |
| MPTr+LapPEBackbone=MPTr, Positional encoding=LapPE2026.06 | 80.8 | |
| SYN-FUSIONfusion_source=Hu et. al2023.08 | 80.5 | |
| GINGraph-level pre-training=Supervised, Node-level pre-training=Attribute Masking2019.05 | 80.3 | |
| UnifiedMolPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 80.3 | |
| GINGraph-level pre-training=Supervised, Node-level pre-training=Infomax2019.05 | 80.1 | |
| MPTr+CycleSEBackbone=MPTr, Positional encoding=CycleSE2026.06 | 80.1 | |
| ChemBERTa-22023.04 | 79.9 | |
| GINGraph-level pre-training=None, Node-level pre-training=EdgePred2019.05 | 79.9 | |
| SSL pre-trainingPre-training strategy=SSL2026.06 | 79.9 | |
| SYN-FUSIONfusion_source=MolCLR2023.08 | 79.8 | |
| MGSSL2026.02 | 79.7 | |
| GINGraph-level pre-training=None, Node-level pre-training=Context Prediction2019.05 | 79.6 | |
| ContextPredPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 79.6 | |
| MPTr+RWSEBackbone=MPTr, Positional encoding=RWSE2026.06 | 79.5 | |
| GINGraph-level pre-training=None, Node-level pre-training=Attribute Masking2019.05 | 79.3 | |
| G-Contextual2023.11 | 79.3 | |
| GraphMVP2023.11 | 79.3 | |
| AttrMaskingPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 79.3 | |
| GraphMVPPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 79.3 | |
| CP2026.02 | 79.2 | |
| MPTrBackbone=MPTr2026.06 | 79.2 | |
| PretrainGNN2024.09 | 79.1 | |
| GINGraph-level pre-training=Supervised, Node-level pre-training=EdgePred2019.05 | 79.1 | |
| MAM2023.11 | 78.9 | |
| ContextPred2023.11 | 78.8 | |
| MGSSL2023.11 | 78.8 | |
| GraphLoG2023.11 | 78.6 | |
| 3D InfoMax2023.11 | 78.6 | |
| AttentiveFP2026.06 | 78.4 | |
| Attr mask2026.02 | 78.3 | |
| GraphCL2026.02 | 78.3 | |
| GraphMAE2023.11 | 78.2 | |
| MAMEncoder=vanilla VQ-VAE2023.11 | 78.2 | |
| MolCLRbackbone=GIN2023.08 | 77.9 | |
| N-Gram RFClassifier=Random Forest2024.09 | 77.9 | |
| GPT-GNN2023.11 | 77.9 | |
| AttrMask2023.11 | 77.8 | |
| Attentive FP2024.09 | 77.3 | |
| EdgePred2023.11 | 77.3 | |
| JOAOPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 77.3 | |
| MAE2026.02 | 77.2 | |
| JOAO2026.02 | 77.2 | |
| GraphGPSPre-training status=None, Pre-training dataset=N/A, Evaluation protocol=Supervised Learning, Split type=Scaffold-split2024.05 | 77 | |
| Hi-GMAE-GPre-training status=None, Pre-training dataset=N/A, Evaluation protocol=Supervised Learning, Split type=Scaffold-split2024.05 | 77 | |
| GINGraph-level pre-training=Supervised, Node-level pre-training=None2019.05 | 76.9 | |
| No Pre-train2026.02 | 76.8 | |
| SchNet2023.04 | 76.6 | |
| AD-GCL2023.11 | 76.6 | |
| Hi-GMAE-FPre-training status=None, Pre-training dataset=N/A, Evaluation protocol=Supervised Learning, Split type=Scaffold-split2024.05 | 76.6 | |
| Infomax2026.02 | 76.3 | |
| GINGraph-level pre-training=None, Node-level pre-training=Infomax2019.05 | 75.9 | |
| InfomaxPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 75.9 | |
| RGCLPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 75.7 | |
| AD-GCLPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 75.6 | |
| GraphCLPre-training status=Pre-trained, Pre-training dataset=ZINC15, Evaluation protocol=Fine-tuning, Split type=Scaffold-split2024.05 | 75.4 | |
| GraphCLPre-training strategy=GraphCL2026.06 | 75.4 | |
| TMCL2023.11 | 75.1 | |
| SimGRACE2023.11 | 74.9 |