Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning
The proposition of non-fullerene acceptors (NFAs) in organic solar cells has made great progress in the raise of power conversion efficiency, and it also broadens the ways for searching and designing new acceptor molecules. In this work, the design of novel NFAs with required properties is performed...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-08-01
|
Series: | International Journal of Molecular Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/1422-0067/22/16/9099 |
_version_ | 1797523500592791552 |
---|---|
author | Shi-Ping Peng Xin-Yu Yang Yi Zhao |
author_facet | Shi-Ping Peng Xin-Yu Yang Yi Zhao |
author_sort | Shi-Ping Peng |
collection | DOAJ |
description | The proposition of non-fullerene acceptors (NFAs) in organic solar cells has made great progress in the raise of power conversion efficiency, and it also broadens the ways for searching and designing new acceptor molecules. In this work, the design of novel NFAs with required properties is performed with the conditional generative model constructed from a convolutional neural network (CNN). The temporal CNN is firstly trained to be a good string-based molecular conditional generative model to directly generate the desired molecules. The reliability of generated molecular properties is then demonstrated by a graph-based prediction model and evaluated with quantum chemical calculations. Specifically, the global attention mechanism is incorporated in the prediction model to pool the extracted information of molecular structures and provide interpretability. By combining the generative and prediction models, thousands of NFAs with required frontier molecular orbital energies are generated. The generated new molecules essentially explore the chemical space and enrich the database of transformation rules for molecular design. The conditional generation model can also be trained to generate the molecules from molecular fragments, and the contribution of molecular fragments to the properties is subsequently predicted by the prediction model. |
first_indexed | 2024-03-10T08:43:53Z |
format | Article |
id | doaj.art-63efd54414e949248b839001731c75d8 |
institution | Directory Open Access Journal |
issn | 1661-6596 1422-0067 |
language | English |
last_indexed | 2024-03-10T08:43:53Z |
publishDate | 2021-08-01 |
publisher | MDPI AG |
record_format | Article |
series | International Journal of Molecular Sciences |
spelling | doaj.art-63efd54414e949248b839001731c75d82023-11-22T08:05:29ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672021-08-012216909910.3390/ijms22169099Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep LearningShi-Ping Peng0Xin-Yu Yang1Yi Zhao2State Key Laboratory for Physical Chemistry of Solid Surfaces, Fujian Provincial Key Lab of Theoretical and Computational Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, ChinaState Key Laboratory for Physical Chemistry of Solid Surfaces, Fujian Provincial Key Lab of Theoretical and Computational Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, ChinaState Key Laboratory for Physical Chemistry of Solid Surfaces, Fujian Provincial Key Lab of Theoretical and Computational Chemistry, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, ChinaThe proposition of non-fullerene acceptors (NFAs) in organic solar cells has made great progress in the raise of power conversion efficiency, and it also broadens the ways for searching and designing new acceptor molecules. In this work, the design of novel NFAs with required properties is performed with the conditional generative model constructed from a convolutional neural network (CNN). The temporal CNN is firstly trained to be a good string-based molecular conditional generative model to directly generate the desired molecules. The reliability of generated molecular properties is then demonstrated by a graph-based prediction model and evaluated with quantum chemical calculations. Specifically, the global attention mechanism is incorporated in the prediction model to pool the extracted information of molecular structures and provide interpretability. By combining the generative and prediction models, thousands of NFAs with required frontier molecular orbital energies are generated. The generated new molecules essentially explore the chemical space and enrich the database of transformation rules for molecular design. The conditional generation model can also be trained to generate the molecules from molecular fragments, and the contribution of molecular fragments to the properties is subsequently predicted by the prediction model.https://www.mdpi.com/1422-0067/22/16/9099non-fullerene acceptorsconvolutional neural networksmolecular generation modelfrontier molecular orbital energieschemical space |
spellingShingle | Shi-Ping Peng Xin-Yu Yang Yi Zhao Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning International Journal of Molecular Sciences non-fullerene acceptors convolutional neural networks molecular generation model frontier molecular orbital energies chemical space |
title | Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning |
title_full | Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning |
title_fullStr | Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning |
title_full_unstemmed | Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning |
title_short | Molecular Conditional Generation and Property Analysis of Non-Fullerene Acceptors with Deep Learning |
title_sort | molecular conditional generation and property analysis of non fullerene acceptors with deep learning |
topic | non-fullerene acceptors convolutional neural networks molecular generation model frontier molecular orbital energies chemical space |
url | https://www.mdpi.com/1422-0067/22/16/9099 |
work_keys_str_mv | AT shipingpeng molecularconditionalgenerationandpropertyanalysisofnonfullereneacceptorswithdeeplearning AT xinyuyang molecularconditionalgenerationandpropertyanalysisofnonfullereneacceptorswithdeeplearning AT yizhao molecularconditionalgenerationandpropertyanalysisofnonfullereneacceptorswithdeeplearning |