PotentialNet for Molecular Property Prediction

Abstract

The arc of drug discovery entails a multiparameter optimization problem spanning vast length scales. They key parameters range from solubility (angstroms) to protein-ligand binding (nanometers) to in vivo toxicity (meters). Through feature learning---instead of feature engineering---deep neural networks promise to outperform both traditional physics-based and knowledge-based machine learning models for predicting molecular properties pertinent to drug discovery. To this end, we present the PotentialNet family of graph convolutions. These models are specifically designed for and achieve state-of-the-art performance for protein-ligand binding affinity. We further validate these deep neural networks by setting new standards of performance in several ligand-based tasks. In parallel, we introduce a new metric, the Regression Enrichment Factor $E{F}_{\chi }^{(R)}$, to measure the early enrichment of computational models for chemical data. Finally, we introduce a cross-validation strategy based on structural homology clustering that can more accurately measure model generalizability, which crucially distinguishes the aims of machine learning for drug discovery from standard machine learning tasks.

arXiv comment13 pages, 5 figures, 8 tables

References in corpus (7)

• The NumPy array: a structure for efficient numerical computation
• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery
• TopologyNet: Topology based deep convolutional neural networks for biomolecular property predictions
• Atomic Convolutional Networks for Predicting Protein-Ligand Binding Affinity