PotentialNet for Molecular Property Prediction

By Evan N. Feinberg, Debnil Sur, Zhenqin Wu, Brooke E. Husic, Huanghao Mai, Yang Li, Saisai Sun, Jianyi Yang, Bharath Ramsundar, Vijay S. Pande. 2018

428 citations recorded by the source corpus.

The arc of drug discovery entails a multiparameter optimization problem spanning vast length scales. The 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 EFχ(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.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology, Graph Learning, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Kenneth Atz
2. Francesca Grisoni
3. Gisbert Schneider
4. Frank No'e
5. K. Müller
6. José Jiménez-Luna
7. Alexandre Tkatchenko
8. Kyle Swanson

In Materials Chemistry

1. Kenneth Atz
2. Frank No'e
3. K. Müller
4. Gisbert Schneider
5. Francesca Grisoni
6. Cecilia Clementi
7. Alexandre Tkatchenko
8. Jian Tang

In Molecular Biology

1. Frank No'e
2. Cecilia Clementi
3. K. Müller
4. Kenneth Atz
5. Francesca Grisoni
6. Alexandre Tkatchenko
7. Gisbert Schneider
8. Kyle Swanson

In Graph Learning

1. Gisbert Schneider
2. Kenneth Atz
3. Jake P. Taylor-King
4. K. Müller
5. Artem Cherkasov
6. Jian Tang
7. Jiejun Xu
8. Michael M. Bronstein

In Computational Chemistry & Drug Discovery

1. Frank No'e
2. K. Müller
3. Alexandre Tkatchenko
4. Kyle Swanson
5. Cecilia Clementi
6. Gisbert Schneider
7. Francesca Grisoni
8. Kenneth Atz

Citations in the corpus

• Energy-based models for atomic-resolution protein conformations (2020, background): “Similarly, Feinberg et al. (2018) achieved state-of-the-art performance on predicting the binding affinity between proteins and small molecules using graph convolutional networks.”
• Utilizing graph machine learning within drug discovery and development (2021, background): “extended previous work on molecular property prediction [104] to include ADME properties [46].”
• Graph convolutional networks: a comprehensive review (2019, background): “PotentialNet [115] first entails graph convolutions over chemical bonds to learn the features of atoms, then entails both bond-based and spatial distancebased propagation and finally conducts graph gathering over the ligand atoms, followed by a fully connected layer for molecular property prediction”
• Machine-learning methods for ligand-protein molecular docking. (2021, methodology): “The first molecular docking method to use graph data was PotentialNet [93].”
• Artificial intelligence in drug discovery: recent advances and future perspectives (2021, background): “Others studies used graph-based approaches in conjunction with distance and angle-based featurizations toward the same goal [10].”
• All SMILES Variational Autoencoder (2019, result): “(a) ZINC250k M ODEL MAE LOG P MAE QED ECFP [53] 0.38 0.045 CVAE [16] 0.15 0.054 CVAE ENC [16] 0.13 0.037 G RAPH C ONV [12] [38] 0.854 P OTENTIAL N ET [14] 0.857 ± 0.006 T OXIC B LEND [71] 0.862 All SMILES 0.871”
• Message passing neural networks for molecular property prediction (2019, background): “Other approaches have explored the use of 3D atomic coordinates to augment the information provided to these models [46, 39, 25, 17, 18].”
• A high quality, industrial data set for binding affinity prediction: performance comparison in different early drug discovery scenarios (2022, methodology): “We further include two 3D-based ML models referred to as atomic convolutional neural network (ACNN) and PotentialNet [7, 18].”
• Machine learning for molecular simulation (2019, background): “ML is also having a big impact in other areas of Chemistry without involving a physics-based model, for example by directly attempting to make predictions of physicochemical or pharmaceutical properties [38, 39, 40, 41], or to designing materials and molecules with certain desirable properties using”