Tox21

Overview edit

Variants: Tox21, Tox21

Associated Benchmarks

This dataset is used in 3 benchmarks:

• Drug Discovery - Metrics: AUC
• Graph Classification - Metrics: ROC-AUC
• Molecular Property Prediction - Metrics: ROC-AUC

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• Perforated Backpropagation: A Neuroscience Inspired Extension to Artificial Neural Networks (2025) -
• Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck (2024) -
• Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns (2023) -
• Structure to Property: Chemical Element Embeddings and a Deep Learning Approach for Accurate Prediction of Chemical Properties (2023) -
• GIT-Mol: A Multi-modal Large Language Model for Molecular Science with Graph, Image, and Text (2023) -
• MolXPT: Wrapping Molecules with Text for Generative Pre-training (2023) -
• Galactica: A Large Language Model for Science (2022) -
• Fine-Tuning Graph Neural Networks via Graph Topology induced Optimal Transport (2022) -
• ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction (2021) -
• Accurate Learning of Graph Representations with Graph Multiset Pooling (2021) -
• Self-Supervised Graph Transformer on Large-Scale Molecular Data (2020) -
• AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data (2020) -
• All SMILES Variational Autoencoder (2019) -
• Strategies for Pre-training Graph Neural Networks (2019) -
• Analyzing Learned Molecular Representations for Property Prediction (2019) -
• Relational Pooling for Graph Representations (2019) -
• N-Gram Graph: Simple Unsupervised Representation for Graphs, with Applications to Molecules (2018) -
• ToxicBlend: Virtual Screening of Toxic Compounds with Ensemble Predictors (2018) -
• Learning Graph-Level Representation for Drug Discovery (2017) -
• Self-Normalizing Neural Networks (2017) -