High-Throughput Screening of Promising Redox-Active Molecules with MolGAT
Abstract
Redox flow batteries (RFBs) have emerged as a promising option for large-scale energy storage, owing to their high energy density, low cost, and environmental benefits. However, the identification of organic compounds with high redox activity, aqueous solubility, stability, and fast redox kinetics is a crucial and challenging step in developing an RFB technology. Density functional theory-based computational materials prediction and screening is a time-consuming and computationally expensive technique, yet it has a high success rate. To speed up the discovery of new materials with desired properties, machine-learning-based models can be trained on large data sets. Graph neural networks (GNNs) are particularly well-suited for non-Euclidean data and can model complex relationships, making them ideal for accelerating the discovery of novel materials. In this study, a GNN-based model called MolGAT was developed to predict the redox potential of organic molecules using molecular structures, atomic properties, and bond attributes. The model was trained on a data set of over 15,000 compounds with redox potentials ranging from –4.11 to 2.56. MolGAT outperformed other GNN variants, such as the Graph Attention Network, Graph Convolution Network, and AttentiveFP models. The trained model was used to screen a vast chemical data set comprising 581,014 molecules, namely OMDB,more »
- Authors:
-
- Department of Physics, College of Natural and Computational Sciences, Addis Ababa University, P.O. Box 1176, Addis Ababa 1176, Ethiopia, Computational Data Science, College of Natural and Computational Sciences, Addis Ababa University, P.O. Box 1176, Addis Ababa 1176, Ethiopia,
- Department of Physics, College of Natural and Computational Sciences, Addis Ababa University, P.O. Box 1176, Addis Ababa 1176, Ethiopia
- The Abdus Salam International Centre for Theoretical Physics(ICTP) Condensed Matter and Statistical Physics Section, 34100 Trieste, Italy,
- Brookhaven National Laboratory, Center for Functional Nanomaterials, Upton New York 11973, United States,
- Center for Environmental Science, College of Natural and Computational Sciences, Addis Ababa University, P.O. Box 1176, Addis Ababa 1176, Ethiopia,
- Publication Date:
- Research Org.:
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF); USDOE Office of Science (SC), Basic Energy Sciences (BES)
- OSTI Identifier:
- 1987931
- Alternate Identifier(s):
- OSTI ID: 1989170; OSTI ID: 1992872
- Report Number(s):
- BNL-224611-2023-JAAM
Journal ID: ISSN 2470-1343
- Grant/Contract Number:
- SC0012704
- Resource Type:
- Published Article
- Journal Name:
- ACS Omega
- Additional Journal Information:
- Journal Name: ACS Omega Journal Volume: 8 Journal Issue: 27; Journal ID: ISSN 2470-1343
- Publisher:
- American Chemical Society
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Drug discovery; Materials; Molecular modeling; Molecules; Redox reactions; 37 INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL CHEMISTRY
Citation Formats
Chaka, Mesfin Diro, Geffe, Chernet Amente, Rodriguez, Alex, Seriani, Nicola, Wu, Qin, and Mekonnen, Yedilfana Setarge. High-Throughput Screening of Promising Redox-Active Molecules with MolGAT. United States: N. p., 2023.
Web. doi:10.1021/acsomega.3c01295.
Chaka, Mesfin Diro, Geffe, Chernet Amente, Rodriguez, Alex, Seriani, Nicola, Wu, Qin, & Mekonnen, Yedilfana Setarge. High-Throughput Screening of Promising Redox-Active Molecules with MolGAT. United States. https://doi.org/10.1021/acsomega.3c01295
Chaka, Mesfin Diro, Geffe, Chernet Amente, Rodriguez, Alex, Seriani, Nicola, Wu, Qin, and Mekonnen, Yedilfana Setarge. Fri .
"High-Throughput Screening of Promising Redox-Active Molecules with MolGAT". United States. https://doi.org/10.1021/acsomega.3c01295.
@article{osti_1987931,
title = {High-Throughput Screening of Promising Redox-Active Molecules with MolGAT},
author = {Chaka, Mesfin Diro and Geffe, Chernet Amente and Rodriguez, Alex and Seriani, Nicola and Wu, Qin and Mekonnen, Yedilfana Setarge},
abstractNote = {Redox flow batteries (RFBs) have emerged as a promising option for large-scale energy storage, owing to their high energy density, low cost, and environmental benefits. However, the identification of organic compounds with high redox activity, aqueous solubility, stability, and fast redox kinetics is a crucial and challenging step in developing an RFB technology. Density functional theory-based computational materials prediction and screening is a time-consuming and computationally expensive technique, yet it has a high success rate. To speed up the discovery of new materials with desired properties, machine-learning-based models can be trained on large data sets. Graph neural networks (GNNs) are particularly well-suited for non-Euclidean data and can model complex relationships, making them ideal for accelerating the discovery of novel materials. In this study, a GNN-based model called MolGAT was developed to predict the redox potential of organic molecules using molecular structures, atomic properties, and bond attributes. The model was trained on a data set of over 15,000 compounds with redox potentials ranging from –4.11 to 2.56. MolGAT outperformed other GNN variants, such as the Graph Attention Network, Graph Convolution Network, and AttentiveFP models. The trained model was used to screen a vast chemical data set comprising 581,014 molecules, namely OMDB, QM9, ZINC, CHEMBL, and DELANEY, and identified 23,467 potential redox-active compounds for use in redox flow batteries. Of those, 20,716 molecules were identified as potential catholytes with predicted redox potentials up to 2.87 V, while 2,751 molecules were deemed potential anolytes with predicted redox potentials as low as –2.88 V. This work demonstrates the capabilities of graph neural networks in condensed matter physics and materials science to screen promising redox-active species for further electronic structure calculations and experimental testing.},
doi = {10.1021/acsomega.3c01295},
journal = {ACS Omega},
number = 27,
volume = 8,
place = {United States},
year = {Fri Jun 30 00:00:00 EDT 2023},
month = {Fri Jun 30 00:00:00 EDT 2023}
}
https://doi.org/10.1021/acsomega.3c01295
Works referenced in this record:
Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science
journal, April 2016
- Agrawal, Ankit; Choudhary, Alok
- APL Materials, Vol. 4, Issue 5
Unified Representation of Molecules and Crystals for Machine Learning
text, January 2017
- Huo, Haoyan; Rupp, Matthias
- arXiv
Crystal Structure Prediction via Deep Learning
journal, June 2018
- Ryan, Kevin; Lengyel, Jeff; Shatruk, Michael
- Journal of the American Chemical Society, Vol. 140, Issue 32
The Generation of a Unique Machine Description for Chemical Structures-A Technique Developed at Chemical Abstracts Service.
journal, May 1965
- Morgan, H. L.
- Journal of Chemical Documentation, Vol. 5, Issue 2
kGCN: a graph-based deep learning framework for chemical structures
journal, May 2020
- Kojima, Ryosuke; Ishida, Shoichi; Ohta, Masateru
- Journal of Cheminformatics, Vol. 12, Issue 1
Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models
journal, February 2021
- Jiang, Dejun; Wu, Zhenxing; Hsieh, Chang-Yu
- Journal of Cheminformatics, Vol. 13, Issue 1
Generative Deep Neural Networks for Inverse Materials Design Using Backpropagation and Active Learning
journal, January 2020
- Chen, Chun‐Teh; Gu, Grace X.
- Advanced Science, Vol. 7, Issue 5
Issues and challenges facing rechargeable lithium batteries
journal, November 2001
- Tarascon, J.-M.; Armand, M.
- Nature, Vol. 414, Issue 6861, p. 359-367
Reverse graph self-attention for target-directed atomic importance estimation
journal, January 2021
- Na, Gyoung S.; Kim, Hyun Woo
- Neural Networks, Vol. 133
Extended-Connectivity Fingerprints
journal, April 2010
- Rogers, David; Hahn, Mathew
- Journal of Chemical Information and Modeling, Vol. 50, Issue 5
Modeling Relational Data with Graph Convolutional Networks
preprint, January 2017
- Schlichtkrull, Michael; Kipf, Thomas N.; Bloem, Peter
- arXiv
Deep Neural Nets as a Method for Quantitative Structure–Activity Relationships
journal, February 2015
- Ma, Junshui; Sheridan, Robert P.; Liaw, Andy
- Journal of Chemical Information and Modeling, Vol. 55, Issue 2
Fast 3D-lithium-ion diffusion and high electronic conductivity of Li2MnSiO4 surfaces for rechargeable lithium-ion batteries
journal, January 2021
- Gurmesa, Gamachis Sakata; Benti, Natei Ermias; Chaka, Mesfin Diro
- RSC Advances, Vol. 11, Issue 16
Communication: The influence of CO 2 poisoning on overvoltages and discharge capacity in non-aqueous Li-Air batteries
journal, March 2014
- Mekonnen, Yedilfana S.; Knudsen, Kristian B.; Mýrdal, Jon S. G.
- The Journal of Chemical Physics, Vol. 140, Issue 12
Organic Electroactive Molecule-Based Electrolytes for Redox Flow Batteries: Status and Challenges of Molecular Design
journal, June 2020
- Zhong, Fangfang; Yang, Minghui; Ding, Mei
- Frontiers in Chemistry, Vol. 8
Accelerating Electrolyte Discovery for Energy Storage with High-Throughput Screening
journal, January 2015
- Cheng, Lei; Assary, Rajeev S.; Qu, Xiaohui
- The Journal of Physical Chemistry Letters, Vol. 6, Issue 2
Fast Graph Representation Learning with PyTorch Geometric
preprint, January 2019
- Fey, Matthias; Lenssen, Jan Eric
- arXiv
Machine learning-based high throughput screening for nitrogen fixation on boron-doped single atom catalysts
journal, January 2020
- Zafari, Mohammad; Kumar, Deepak; Umer, Muhammad
- Journal of Materials Chemistry A, Vol. 8, Issue 10
Inverse design of glass structure with deep graph neural networks
journal, September 2021
- Wang, Qi; Zhang, Longfei
- Nature Communications, Vol. 12, Issue 1
Organic materials database: An open-access online database for data mining
journal, February 2017
- Borysov, Stanislav S.; Geilhufe, R. Matthias; Balatsky, Alexander V.
- PLOS ONE, Vol. 12, Issue 2
ZINC − A Free Database of Commercially Available Compounds for Virtual Screening
journal, December 2004
- Irwin, John J.; Shoichet, Brian K.
- Journal of Chemical Information and Modeling, Vol. 45, Issue 1
Electrical Energy Storage for the Grid: A Battery of Choices
journal, November 2011
- Dunn, B.; Kamath, H.; Tarascon, J. -M.
- Science, Vol. 334, Issue 6058
Li-ion battery materials: present and future
journal, June 2015
- Nitta, Naoki; Wu, Feixiang; Lee, Jung Tae
- Materials Today, Vol. 18, Issue 5
Predicting materials properties without crystal structure: deep representation learning from stoichiometry
journal, December 2020
- Goodall, Rhys E. A.; Lee, Alpha A.
- Nature Communications, Vol. 11, Issue 1
Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
journal, April 2019
- Chen, Chi; Ye, Weike; Zuo, Yunxing
- Chemistry of Materials, Vol. 31, Issue 9
Exploiting Edge Features in Graph Neural Networks
preprint, January 2018
- Gong, Liyu; Cheng, Qiang
- arXiv
Sodium-ion diffusion studies of the cathode–electrolyte interfaces (NaxO2@Na2CO3, x=1 and 2) and discharge products of non-aqueous rechargeable sodium–air batteries
journal, January 2022
- Benti, Natei Ermias; Gurmesa, Gamachis Sakata; Geffe, Chernet Amente
- Journal of Materials Chemistry A, Vol. 10, Issue 15
Machine Learning in Materials Science
book, January 2016
- Mueller, Tim; Kusne, Aaron Gilad; Ramprasad, Rampi
- Reviews in Computational Chemistry, Vol. 29
Inverse Design of Solid-State Materials via a Continuous Representation
journal, November 2019
- Noh, Juhwan; Kim, Jaehoon; Stein, Helge S.
- Matter, Vol. 1, Issue 5
GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
preprint, January 2020
- Cai, Tianle; Luo, Shengjie; Xu, Keyulu
- arXiv
Learning from the Harvard Clean Energy Project: The Use of Neural Networks to Accelerate Materials Discovery
journal, September 2015
- Pyzer-Knapp, Edward O.; Li, Kewei; Aspuru-Guzik, Alan
- Advanced Functional Materials, Vol. 25, Issue 41
Cryo-EM structure of a SARS-CoV-2 omicron spike protein ectodomain
journal, March 2022
- Ye, Gang; Liu, Bin; Li, Fang
- Nature Communications, Vol. 13, Issue 1
ESOL: Estimating Aqueous Solubility Directly from Molecular Structure
journal, May 2004
- Delaney, John S.
- Journal of Chemical Information and Computer Sciences, Vol. 44, Issue 3
Semi-Supervised Classification with Graph Convolutional Networks
preprint, January 2016
- Kipf, Thomas N.; Welling, Max
- arXiv
Molecular graph convolutions: moving beyond fingerprints
journal, August 2016
- Kearnes, Steven; McCloskey, Kevin; Berndl, Marc
- Journal of Computer-Aided Molecular Design, Vol. 30, Issue 8
Graph Neural Networks: A Review of Methods and Applications
preprint, January 2018
- Zhou, Jie; Cui, Ganqu; Hu, Shengding
- arXiv
The ChEMBL database in 2017
journal, November 2016
- Gaulton, Anna; Hersey, Anne; Nowotka, Michał
- Nucleic Acids Research, Vol. 45, Issue D1
Machine learning in materials science
journal, August 2019
- Wei, Jing; Chu, Xuan; Sun, Xiang‐Yu
- InfoMat, Vol. 1, Issue 3
Redox flow batteries: Status and perspective towards sustainable stationary energy storage
journal, January 2021
- Sánchez-Díez, Eduardo; Ventosa, Edgar; Guarnieri, Massimo
- Journal of Power Sources, Vol. 481
Computational design of molecules for an all-quinone redox flow battery
journal, January 2015
- Er, Süleyman; Suh, Changwon; Marshak, Michael P.
- Chemical Science, Vol. 6, Issue 2, p. 885-893
Thermodynamic and Kinetic Limitations for Peroxide and Superoxide Formation in Na–O 2 Batteries
journal, July 2018
- Mekonnen, Yedilfana S.; Christensen, Rune; Garcia-Lastra, Juan M.
- The Journal of Physical Chemistry Letters, Vol. 9, Issue 15
Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learning
preprint, January 2018
- Li, Qimai; Han, Zhichao; Wu, Xiao-Ming
- arXiv
Development of high-energy non-aqueous lithium-sulfur batteries via redox-active interlayer strategy
journal, August 2022
- Lee, Byong-June; Zhao, Chen; Yu, Jeong-Hoon
- Nature Communications, Vol. 13, Issue 1
A machine learning–based classification approach for phase diagram prediction
journal, March 2022
- Deffrennes, Guillaume; Terayama, Kei; Abe, Taichi
- Materials & Design, Vol. 215
Effects of Functional Groups in Redox-Active Organic Molecules: A High-Throughput Screening Approach
journal, December 2016
- Pelzer, Kenley M.; Cheng, Lei; Curtiss, Larry A.
- The Journal of Physical Chemistry C, Vol. 121, Issue 1
Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD)
journal, September 2013
- Saal, James E.; Kirklin, Scott; Aykol, Muratahan
- JOM, Vol. 65, Issue 11
Rational Design of Biaxial Tensile Strain for Boosting Electronic and Ionic Conductivities of Na
2
MnSiO
4
for Rechargeable Sodium‐Ion Batteries
journal, June 2022
- Sakata Gurmesa, Gamachis; Teshome, Tamiru; Ermias Benti, Natei
- ChemistryOpen, Vol. 11, Issue 6
Convolutional Networks on Graphs for Learning Molecular Fingerprints
preprint, January 2015
- Duvenaud, David; Maclaurin, Dougal; Aguilera-Iparraguirre, Jorge
- arXiv
A graph-convolutional neural network model for the prediction of chemical reactivity
journal, January 2019
- Coley, Connor W.; Jin, Wengong; Rogers, Luke
- Chemical Science, Vol. 10, Issue 2
A high throughput molecular screening for organic electronics via machine learning: present status and perspective
journal, December 2019
- Saeki, Akinori; Kranthiraja, Kakaraparthi
- Japanese Journal of Applied Physics, Vol. 59, Issue SD
Molecular structure–redox potential relationship for organic electrode materials: density functional theory–Machine learning approach
journal, September 2020
- Allam, O.; Kuramshin, R.; Stoichev, Z.
- Materials Today Energy, Vol. 17
Introduction
book, January 2020
- Hamilton, William L.
- Synthesis Lectures on Artificial Intelligence and Machine Learning
Machine Learning-Assisted High-Throughput Virtual Screening for On-Demand Customization of Advanced Energetic Materials
journal, March 2022
- Song, Siwei; Wang, Yi; Chen, Fang
- Engineering, Vol. 10
Predicting Materials Properties with Little Data Using Shotgun Transfer Learning
journal, September 2019
- Yamada, Hironao; Liu, Chang; Wu, Stephen
- ACS Central Science, Vol. 5, Issue 10
Pushing the Boundaries of Molecular Representation for Drug Discovery with the Graph Attention Mechanism
journal, August 2019
- Xiong, Zhaoping; Wang, Dingyan; Liu, Xiaohong
- Journal of Medicinal Chemistry, Vol. 63, Issue 16
Inverse design of nanoporous crystalline reticular materials with deep generative models
journal, January 2021
- Yao, Zhenpeng; Sánchez-Lengeling, Benjamín; Bobbitt, N. Scott
- Nature Machine Intelligence, Vol. 3, Issue 1
Deep-learning-based inverse design model for intelligent discovery of organic molecules
journal, December 2018
- Kim, Kyungdoc; Kang, Seokho; Yoo, Jiho
- npj Computational Materials, Vol. 4, Issue 1
Quantum chemistry structures and properties of 134 kilo molecules
journal, August 2014
- Ramakrishnan, Raghunathan; Dral, Pavlo O.; Rupp, Matthias
- Scientific Data, Vol. 1, Issue 1
Inverse Design of Materials by Machine Learning
journal, February 2022
- Wang, Jia; Wang, Yingxue; Chen, Yanan
- Materials, Vol. 15, Issue 5
Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
journal, April 2018
- Xie, Tian; Grossman, Jeffrey C.
- Physical Review Letters, Vol. 120, Issue 14
Machine Learning-Driven High-Throughput Screening of Alloy-Based Catalysts for Selective CO2 Hydrogenation to Methanol
journal, November 2021
- Roy, Diptendu; Mandal, Shyama Charan; Pathak, Biswarup
- ACS Applied Materials & Interfaces, Vol. 13, Issue 47
Planning chemical syntheses with deep neural networks and symbolic AI
journal, March 2018
- Segler, Marwin H. S.; Preuss, Mike; Waller, Mark P.
- Nature, Vol. 555, Issue 7698
A metal-free organic–inorganic aqueous flow battery
journal, January 2014
- Huskinson, Brian; Marshak, Michael P.; Suh, Changwon
- Nature, Vol. 505, Issue 7482, p. 195-198