MoleculeNet
Lipophilicity
Structure
4,200Graphs
27Nodes
30Edges
1Classes
9Node Feature Dim
3Edge Feature Dim
Coverage
122Results
1Variants
14Papers
- Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction
- MolGA: Molecular Graph Adaptation with Pre-trained 2D Graph Encoder MolGA: Molecular Graph Adaptation with Pre-trained 2D Graph Encoder
- AAAI 2024: Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck AAAI 2024
- NeurIPS 2023: SE(3)-Invariant Multiparameter Persistent Homology for Chiral-Sensitive Molecular Property Prediction SE(3)-Invariant Multiparameter Persistent Homology for Chiral-Sensitive Molecular Property Prediction NeurIPS 2023
- ICML 2023: Graph Neural Networks with a Distribution of Parametrized Graphs Graph Neural Networks with a Distribution of Parametrized Graphs ICML 2023
- Motif-aware Attribute Masking for Molecular Graph Pre-training Motif-aware Attribute Masking for Molecular Graph Pre-training
- ICML 2023: Graph Positional and Structural Encoder Graph Positional and Structural Encoder ICML 2023
- 3D Molecular Geometry Analysis with 2D Graphs 3D Molecular Geometry Analysis with 2D Graphs
- Application of Graph Neural Networks and graph descriptors for graph classification Application of Graph Neural Networks and graph descriptors for graph classification
- Improving Molecular Pretraining with Complementary Featurizations Improving Molecular Pretraining with Complementary Featurizations
- AAAI 2021: GeomGCL: Geometric Graph Contrastive Learning for Molecular Property Prediction GeomGCL: Geometric Graph Contrastive Learning for Molecular Property Prediction AAAI 2021
- Optimal Transport Graph Neural Networks Optimal Transport Graph Neural Networks
- ICLR 2020: Memory-Based Graph Networks Memory-Based Graph Networks ICLR 2020
- MoleculeNet MoleculeNet
Graph regression Octanol/water distribution coefficient (logD) regression on 4,200 molecules. SMILES and labels are fixed; molecular graph featurization is implementation-defined. Random and scaffold splits are not comparable. Evaluated by RMSE.
Dataset details
Task
Octanol/water distribution coefficient (logD) regression on 4,200 molecules. The primary catalog metric is RMSE.
Data
Octanol/water distribution coefficient (logD) regression on 4,200 molecules.
Features
MoleculeNet distributes SMILES and targets, not one mandatory graph featurization. A graph implementation normally derives atom and bond fields with RDKit or DeepChem; results depend on featurizer, salt handling, stereochemistry, and invalid-molecule policy.
Splits and evaluation
The catalog records these protocols or variants: Scaffold split. Evaluation uses RMSE (lower is better). Exact masks or folds must come from the cited release.
Quirks and pitfalls
Small samples, imbalance, missing assay labels, duplicate compounds, and scaffold leakage are common. Missing labels are not negatives. Always report split algorithm, featurizer, metric aggregation, and dataset version; random and scaffold results are not comparable.
Sources
SOTA Progression
new SOTA
matches SOTA at that time
other proposed result
other baseline result
Leaderboard
| # | Model | RMSE ↓ | Paper |
|---|---|---|---|
| 1 | MemGNN | 0.5560
±
0.0230 | |
| 2 | ProtoS-L2 | 0.5800
±
0.0160 | |
| 3 | ProtoW-L2 | 0.6040
±
0.0140 | |
| 4 | ProtoW-Dot | 0.6290
±
0.0150 | |
| 5 | GPS + GPSE | 0.6430 | |
| 6 | D-MPNN | 0.6460
±
0.0410 | |
| 7 | GCN | 0.6550
±
0.0360 | |
| 8 | GC | 0.6550
±
0.0360 | |
| 9 | GIN | 0.6580
±
0.0190 | |
| 10 | GeomGCL | 0.6610 | |
| 11 | GPS+LapPE | 0.6620 | |
| 12 | GPS + RWSE | 0.6710 | |
| 13 | Node-MPN | 0.6720
±
0.0510 | |
| 14 | D-MPNN+TopK Pool | 0.6750
±
0.0800 | |
| 15 | GeomMPNN | 0.6820 | |
| 16 | 3D Infomax | 0.6950
±
0.0120 | |
| 17 | ContextPred | 0.7020
±
0.0200 | |
| 18 | CP | 0.7020 | |
| 19 | ECFP-4 + RF | 0.7060
±
0.0110 | |
| 20 | MOCO | 0.7070
±
0.0010 | |
| 21 | JOAO | 0.7080
±
0.0070 | |
| 22 | GraphConv | 0.7120
±
0.0490 | |
| 23 | GraphCL | 0.7140
±
0.0110 | |
| 24 | GCN + JK concat | 0.7150 | |
| 25 | Weave | 0.7150
±
0.0350 | |
| 26 | GraphMVP | 0.7180
±
0.0160 | |
| 27 | MPNN | 0.7190
±
0.0310 | |
| 28 | D-MPNN+SAG Pool | 0.7200
±
0.0390 | |
| 29 | GCN + JK LSTM | 0.7210 | |
| 30 | GraphSAGE + JK LSTM | 0.7260 | |
| 31 | AttrMask | 0.7300
±
0.0040 | |
| 32 | AM | 0.7300 | |
| 33 | GraphSAGE + JK concat | 0.7320 | |
| 34 | GAP-EMPNN | 0.7320
±
0.0050 | |
| 35 | GAT + JK max | 0.7340 | |
| 36 | GIN + JK LSTM | 0.7350 | |
| 37 | --- | 0.7360
±
0.0060 | |
| 38 | AttentiveFP | 0.7380 | |
| 39 | SE(3)-I MPPH | 0.7380
±
0.0250 | |
| 40 | DimeNet++ | 0.7420 | |
| 41 | EM-GCN | 0.7430
±
0.0130 | |
| 42 | GIN + JK max | 0.7470 | |
| 43 | GCN | 0.7600
±
0.0090 | |
| 44 | CoMPT | 0.7610 | |
| 45 | S-CGIB | 0.7620
±
0.0420 | |
| 46 | AttentiveFP | 0.7620
±
0.0220 | |
| 47 | GCN + JK max | 0.7630 | |
| 48 | SimSGT | 0.7710
±
0.0410 | |
| 49 | MoAMa | 0.7720
±
0.0300 | |
| 50 | MoleBlend | 0.7740
±
0.0120 |
SOTA performance
set a new SOTA at the time
matched the SOTA at the time