Evolutionary design of molecules based on deep learning and a genetic algorithm

Abstract Evolutionary design has gained significant attention as a useful tool to accelerate the design process by automatically modifying molecular structures to obtain molecules with the target properties. However, its methodology presents a practical challenge—devising a way in which to rapidly e...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Youngchun Kwon, Seokho Kang, Youn-Suk Choi, Inkoo Kim
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/3efa6c8a215f4b9ab91a246a436434f9
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:3efa6c8a215f4b9ab91a246a436434f9
record_format	dspace
spelling	oai:doaj.org-article:3efa6c8a215f4b9ab91a246a436434f92021-12-02T16:34:59ZEvolutionary design of molecules based on deep learning and a genetic algorithm10.1038/s41598-021-96812-82045-2322https://doaj.org/article/3efa6c8a215f4b9ab91a246a436434f92021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96812-8https://doaj.org/toc/2045-2322Abstract Evolutionary design has gained significant attention as a useful tool to accelerate the design process by automatically modifying molecular structures to obtain molecules with the target properties. However, its methodology presents a practical challenge—devising a way in which to rapidly evolve molecules while maintaining their chemical validity. In this study, we address this limitation by developing an evolutionary design method. The method employs deep learning models to extract the inherent knowledge from a database of materials and is used to effectively guide the evolutionary design. In the proposed method, the Morgan fingerprint vectors of seed molecules are evolved using the techniques of mutation and crossover within the genetic algorithm. Then, a recurrent neural network is used to reconstruct the final fingerprints into actual molecular structures while maintaining their chemical validity. The use of deep neural network models to predict the properties of these molecules enabled more versatile and efficient molecular evaluations to be conducted by using the proposed method repeatedly. Four design tasks were performed to modify the light-absorbing wavelengths of organic molecules from the PubChem library.Youngchun KwonSeokho KangYoun-Suk ChoiInkoo KimNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Youngchun Kwon Seokho Kang Youn-Suk Choi Inkoo Kim Evolutionary design of molecules based on deep learning and a genetic algorithm
description	Abstract Evolutionary design has gained significant attention as a useful tool to accelerate the design process by automatically modifying molecular structures to obtain molecules with the target properties. However, its methodology presents a practical challenge—devising a way in which to rapidly evolve molecules while maintaining their chemical validity. In this study, we address this limitation by developing an evolutionary design method. The method employs deep learning models to extract the inherent knowledge from a database of materials and is used to effectively guide the evolutionary design. In the proposed method, the Morgan fingerprint vectors of seed molecules are evolved using the techniques of mutation and crossover within the genetic algorithm. Then, a recurrent neural network is used to reconstruct the final fingerprints into actual molecular structures while maintaining their chemical validity. The use of deep neural network models to predict the properties of these molecules enabled more versatile and efficient molecular evaluations to be conducted by using the proposed method repeatedly. Four design tasks were performed to modify the light-absorbing wavelengths of organic molecules from the PubChem library.
format	article
author	Youngchun Kwon Seokho Kang Youn-Suk Choi Inkoo Kim
author_facet	Youngchun Kwon Seokho Kang Youn-Suk Choi Inkoo Kim
author_sort	Youngchun Kwon
title	Evolutionary design of molecules based on deep learning and a genetic algorithm
title_short	Evolutionary design of molecules based on deep learning and a genetic algorithm
title_full	Evolutionary design of molecules based on deep learning and a genetic algorithm
title_fullStr	Evolutionary design of molecules based on deep learning and a genetic algorithm
title_full_unstemmed	Evolutionary design of molecules based on deep learning and a genetic algorithm
title_sort	evolutionary design of molecules based on deep learning and a genetic algorithm
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/3efa6c8a215f4b9ab91a246a436434f9
work_keys_str_mv	AT youngchunkwon evolutionarydesignofmoleculesbasedondeeplearningandageneticalgorithm AT seokhokang evolutionarydesignofmoleculesbasedondeeplearningandageneticalgorithm AT younsukchoi evolutionarydesignofmoleculesbasedondeeplearningandageneticalgorithm AT inkookim evolutionarydesignofmoleculesbasedondeeplearningandageneticalgorithm
_version_	1718383746987589632

Evolutionary design of molecules based on deep learning and a genetic algorithm

Ejemplares similares