Papers › Self-Supervised Graph Transformer on Large-Scale Molecular Data

Self-Supervised Graph Transformer on Large-Scale Molecular Data

18 Jun 2020NeurIPS 2020 12arXiv:2007.02835archive 2025-07-28

Yu Rong, Yatao Bian, Tingyang Xu, Weiyang Xie, Ying WEI, Wenbing Huang, Junzhou Huang

How to obtain informative representations of molecules is a crucial prerequisite in AI-driven drug design and discovery. Recent researches abstract molecules as graphs and employ Graph Neural Networks (GNNs) for molecular representation learning. Nevertheless, two issues impede the usage of GNNs in real scenarios: (1) insufficient labeled molecules for supervised training; (2) poor generalization capability to new-synthesized molecules. To address them both, we propose a novel framework, GROVER, which stands for Graph Representation frOm self-superVised mEssage passing tRansformer. With carefully designed self-supervised tasks in node-, edge- and graph-level, GROVER can learn rich structural and semantic information of molecules from enormous unlabelled molecular data. Rather, to encode such complex information, GROVER integrates Message Passing Networks into the Transformer-style architecture to deliver a class of more expressive encoders of molecules. The flexibility of GROVER allows it to be trained efficiently on large-scale molecular dataset without requiring any supervision, thus being immunized to the two issues mentioned above. We pre-train GROVER with 100 million parameters on 10 million unlabelled molecules -- the biggest GNN and the largest training dataset in molecular representation learning. We then leverage the pre-trained GROVER for molecular property prediction followed by task-specific fine-tuning, where we observe a huge improvement (more than 6% on average) from current state-of-the-art methods on 11 challenging benchmarks. The insights we gained are that well-designed self-supervision losses and largely-expressive pre-trained models enjoy the significant potential on performance boosting.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

tencent-ailab/grover officialpytorchNOASSERTION report
dengjianyuan/respite_mpp mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Drug DesignMolecular Property PredictionProperty PredictionRepresentation Learningmolecular representation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Molecular Property Prediction BACE GROVER (base) ROC-AUC 82.6 #9 of 20 Archive leaderboard report
Molecular Property Prediction BACE GROVER (large) ROC-AUC 81.0 #10 of 20 Archive leaderboard report
Molecular Property Prediction BBBP GROVER (base) ROC-AUC 70.0 #20 of 29 Archive leaderboard report
Molecular Property Prediction BBBP GROVER (large) ROC-AUC 69.5 #22 of 29 Archive leaderboard report
Molecular Property Prediction FreeSolv GROVER (base) RMSE 2.176 #18 of 22 Archive leaderboard report
Molecular Property Prediction FreeSolv GROVER (large) RMSE 2.272 #19 of 22 Archive leaderboard report
Molecular Property Prediction Lipophilicity GROVER (base) RMSE 0.817 #11 of 13 Archive leaderboard report
Molecular Property Prediction Lipophilicity GROVER (large) RMSE 0.823 #12 of 13 Archive leaderboard report
Molecular Property Prediction QM7 GROVER (large) MAE 92.0 #4 of 8 Archive leaderboard report
Molecular Property Prediction QM7 GROVER (base) MAE 94.5 #6 of 8 Archive leaderboard report
Molecular Property Prediction QM8 GROVER (base) MAE 0.0218 #6 of 8 Archive leaderboard report
Molecular Property Prediction QM8 GROVER (large) MAE 0.0224 #7 of 8 Archive leaderboard report
Molecular Property Prediction QM9 GROVER (base) MAE 0.00984 #6 of 8 Archive leaderboard report
Molecular Property Prediction QM9 GROVER (large) MAE 0.00986 #7 of 8 Archive leaderboard report
Molecular Property Prediction SIDER GROVER (large) ROC-AUC 65.4 #9 of 19 Archive leaderboard report
Molecular Property Prediction SIDER GROVER (base) ROC-AUC 64.8 #10 of 19 Archive leaderboard report
Molecular Property Prediction Tox21 GROVER (base) ROC-AUC 74.3 #14 of 20 Archive leaderboard report
Molecular Property Prediction Tox21 GROVER (large) ROC-AUC 73.5 #15 of 20 Archive leaderboard report
Molecular Property Prediction ToxCast GROVER (base) ROC-AUC 65.4 #7 of 8 Archive leaderboard report
Molecular Property Prediction ToxCast GROVER (large) ROC-AUC 65.3 #8 of 8 Archive leaderboard report
Molecular Property Prediction clintox GROVER (base) Molecules (M) 11 #12 of 20 Archive leaderboard report
Molecular Property Prediction clintox GROVER (base) ROC-AUC 81.2 #12 of 20 Archive leaderboard report
Molecular Property Prediction clintox GROVER (large) Molecules (M) 11 #16 of 20 Archive leaderboard report
Molecular Property Prediction clintox GROVER (large) ROC-AUC 76.2 #16 of 20 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections