MoleculeNet: a benchmark for molecular machine learning

Wu, Zhenqin; Ramsundar, Bharath; Feinberg, Evan N.; Gomes, Joseph; Geniesse, Caleb; Pappu, Aneesh; Leswing, Karl; Pande, Vijay S.

doi:10.1039/c7sc02664a

Cited by 1,928 publications

(2,654 citation statements)

References 78 publications

Supporting

Mentioning

2,457

Contrasting

Unclassified

Order By: Relevance

“…To date there have been relatively few studies that have made comparisons of deep learning to the wide array of classical machine learning methods or have discussed this methods application in pharmaceutical research 41, 110, 111 or even used the models for actual predictions for ongoing projects. This study therefore fills a void related to drug discovery applications of these methods.…”

Section: Discussionmentioning

confidence: 99%

“…Deep learning with multitask learning 39 slightly outperformed the closest consensus ANN method 40 across nuclear receptor and stress response datasets. Most recently, one group has suggested some datasets for molecular machine learning and used these for comparison with selected machine learning methods 41 . A second group has assessed several machine learning methods with 7 ChEMBL datasets but only focused on a single metric to assess performance 42 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of Deep Learning With Multiple Machine Learning Methods and Metrics Using Diverse Drug Discovery Data Sets

Korotcov¹,

Tkachenko²,

et al. 2017

View full text Add to dashboard Cite

Machine learning methods have been applied to many datasets in pharmaceutical research for several decades. The relative ease and availability of fingerprint type molecular descriptors paired with Bayesian methods resulted in the widespread use of this approach for a diverse array of endpoints relevant to drug discovery. Deep learning is the latest machine learning algorithm attracting attention for many of pharmaceutical applications from docking to virtual screening. Deep learning is based on an artificial neural network with multiple hidden layers and has found considerable traction for many artificial intelligence applications. We have previously suggested the need for a comparison of different machine learning methods with deep learning across an array of varying datasets that is applicable to pharmaceutical research. Endpoints relevant to pharmaceutical research include absorption, distribution, metabolism, excretion and toxicity (ADME/Tox) properties, as well as activity against pathogens and drug discovery datasets. In this study, we have used datasets for solubility, probe-likeness, hERG, KCNQ1, bubonic plague, Chagas, tuberculosis and malaria to compare different machine learning methods using FCFP6 fingerprints. These datasets represent whole cell screens, individual proteins, physicochemical properties as well as a dataset with a complex endpoint. Our aim was to assess whether deep learning offered any improvement in testing when assessed using an array of metrics including AUC, F1 score, Cohen’s kappa, Matthews correlation coefficient and others. Based on ranked normalized scores for the metrics or datasets Deep Neural Networks (DNN) ranked higher than SVM, which in turn was ranked higher than all the other machine learning methods. Visualizing these properties for training and test sets using radar type plots indicates when models are inferior or perhaps over trained. These results also suggest the need for assessing deep learning further using multiple metrics with much larger scale comparisons, prospective testing as well as assessment of different fingerprints and DNN architectures beyond those used.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Comparison of Deep Learning With Multiple Machine Learning Methods and Metrics Using Diverse Drug Discovery Data Sets

Korotcov¹,

Tkachenko²,

et al. 2017

View full text Add to dashboard Cite

show abstract

“…More recently, graph networks further generalize the GCNN by introducing global attributes in addition to atom and bond attributes, and allowing information flow among all three levels of quantities . Graph‐based deep learning models have shown remarkable performance in molecular and crystal property predictions compared to other ML models …”

Section: Model Selection and Trainingmentioning

confidence: 99%

A Critical Review of Machine Learning of Energy Materials

Chen

Zuo

et al. 2020

Advanced Energy Materials

395

281

View full text Add to dashboard Cite

Machine learning (ML) is rapidly revolutionizing many fields and is starting to change landscapes for physics and chemistry. With its ability to solve complex tasks autonomously, ML is being exploited as a radically new way to help find material correlations, understand materials chemistry, and accelerate the discovery of materials. Here, an in‐depth review of the application of ML to energy materials, including rechargeable alkali‐ion batteries, photovoltaics, catalysts, thermoelectrics, piezoelectrics, and superconductors, is presented. A conceptual framework is first provided for ML in materials science, with a broad overview of different ML techniques as well as best practices. This is followed by a critical discussion of how ML is applied in energy materials. This review is concluded with the perspectives on major challenges and opportunities in this exciting field.

show abstract

“…As a data‐driven method, machine learning can bypass the solution of complex equations (eg, the Kohn‐Sham equation or Schrödinger equation) to determine the properties that are related to the energy, geometry, and curvature of the potential energy surfaces of molecules . A group developed a model that is based on a deep tensor neural network for predicting atomic energies and local chemical potentials in molecules, reliable isomer energies, and molecules with peculiar electronic structures, thereby resulting in insights into quantum‐mechanical observables of molecular systems .…”

Section: Applicationsmentioning

confidence: 99%

Machine learning in materials science

Wei

Chu

Sun

et al. 2019

InfoMat

564

298

View full text Add to dashboard Cite

Traditional methods of discovering new materials, such as the empirical trial and error method and the density functional theory (DFT)‐based method, are unable to keep pace with the development of materials science today due to their long development cycles, low efficiency, and high costs. Accordingly, due to its low computational cost and short development cycle, machine learning is coupled with powerful data processing and high prediction performance and is being widely used in material detection, material analysis, and material design. In this article, we discuss the basic operational procedures in analyzing material properties via machine learning, summarize recent applications of machine learning algorithms to several mature fields in materials science, and discuss the improvements that are required for wide‐ranging application.

show abstract

MoleculeNet: a benchmark for molecular machine learning

Abstract: A large scale benchmark for molecular machine learning consisting of multiple public datasets, metrics, featurizations and learning algorithms.

Cited by 1,928 publications

References 78 publications

Comparison of Deep Learning With Multiple Machine Learning Methods and Metrics Using Diverse Drug Discovery Data Sets

Comparison of Deep Learning With Multiple Machine Learning Methods and Metrics Using Diverse Drug Discovery Data Sets

A Critical Review of Machine Learning of Energy Materials

Machine learning in materials science

Contact Info

Product

Resources

About