MoleculeNet: A Benchmark for Molecular Machine Learning

Abstract

Molecular machine learning has been maturing rapidly over the last few years. Improved methods and the presence of larger datasets have enabled machine learning algorithms to make increasingly accurate predictions about molecular properties. However, algorithmic progress has been limited due to the lack of a standard benchmark to compare the efficacy of proposed methods; most new algorithms are benchmarked on different datasets making it challenging to gauge the quality of proposed methods. This work introduces MoleculeNet, a large scale benchmark for molecular machine learning. MoleculeNet curates multiple public datasets, establishes metrics for evaluation, and offers high quality open-source implementations of multiple previously proposed molecular featurization and learning algorithms (released as part of the DeepChem open source library). MoleculeNet benchmarks demonstrate that learnable representations are powerful tools for molecular machine learning and broadly offer the best performance. However, this result comes with caveats. Learnable representations still struggle to deal with complex tasks under data scarcity and highly imbalanced classification. For quantum mechanical and biophysical datasets, the use of physics-aware featurizations can be more important than choice of particular learning algorithm.

References in corpus (4)

• Neural Message Passing for Quantum Chemistry
• Going Deeper with Convolutions
• Massively Multitask Networks for Drug Discovery
• Low Data Drug Discovery with One-shot Learning

Cited by in corpus (20)

• Chemception: A Deep Neural Network with Minimal Chemistry Knowledge Matches the Performance of Expert-developed QSAR/QSPR Models
• Atomic Convolutional Networks for Predicting Protein-Ligand Binding Affinity
• Adaptive Graph Convolutional Neural Networks
• 3D Infomax improves GNNs for Molecular Property Prediction
• PubChemQC PM6: A dataset of 221 million molecules with optimized molecular geometries and electronic properties
• Learning Graph-Level Representation for Drug Discovery
• GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
• SMILES2Vec: An Interpretable General-Purpose Deep Neural Network for Predicting Chemical Properties
• Multi-Objective De Novo Drug Design with Conditional Graph Generative Model
• Active and transfer learning with partially Bayesian neural networks for materials and chemicals
• Learning Graph While Training: An Evolving Graph Convolutional Neural Network
• Scaling Graph Neural Networks to Large Proteins
• Structure-Enhanced Meta-Learning For Few-Shot Graph Classification
• Artificial Intelligence based Autonomous Molecular Design for Medical Therapeutic: A Perspective
• Flexible dual-branched message passing neural network for quantum mechanical property prediction with molecular conformation
• GABO: Graph Augmentations with Bi-level Optimization
• Reprogramming Language Models for Molecular Representation Learning
• LSP : Acceleration and Regularization of Graph Neural Networks via Locality Sensitive Pruning of Graphs
• Topological Bayesian Optimization with Persistence Diagrams
• Symmetry-driven graph neural networks