Deep scaffold hopping with multimodal transformer neural networks

Abstract Scaffold hopping is a central task of modern medicinal chemistry for rational drug design, which aims to design molecules of novel scaffolds sharing similar target biological activities toward known hit molecules. Traditionally, scaffolding hopping depends on searching databases of availabl...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Shuangjia Zheng, Zengrong Lei, Haitao Ai, Hongming Chen, Daiguo Deng, Yuedong Yang
Formato:	article
Lenguaje:	EN
Publicado:	BMC 2021
Materias:	Deep learning Drug design Scaffold hopping Molecular optimization Transformer neural network Information technology T58.5-58.64 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/c9a4a58e594a43799741bcdab9961ad1
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c9a4a58e594a43799741bcdab9961ad1
record_format	dspace
spelling	oai:doaj.org-article:c9a4a58e594a43799741bcdab9961ad12021-11-14T12:33:32ZDeep scaffold hopping with multimodal transformer neural networks10.1186/s13321-021-00565-51758-2946https://doaj.org/article/c9a4a58e594a43799741bcdab9961ad12021-11-01T00:00:00Zhttps://doi.org/10.1186/s13321-021-00565-5https://doaj.org/toc/1758-2946Abstract Scaffold hopping is a central task of modern medicinal chemistry for rational drug design, which aims to design molecules of novel scaffolds sharing similar target biological activities toward known hit molecules. Traditionally, scaffolding hopping depends on searching databases of available compounds that can't exploit vast chemical space. In this study, we have re-formulated this task as a supervised molecule-to-molecule translation to generate hopped molecules novel in 2D structure but similar in 3D structure, as inspired by the fact that candidate compounds bind with their targets through 3D conformations. To efficiently train the model, we curated over 50 thousand pairs of molecules with increased bioactivity, similar 3D structure, but different 2D structure from public bioactivity database, which spanned 40 kinases commonly investigated by medicinal chemists. Moreover, we have designed a multimodal molecular transformer architecture by integrating molecular 3D conformer through a spatial graph neural network and protein sequence information through Transformer. The trained DeepHop model was shown able to generate around 70% molecules having improved bioactivity together with high 3D similarity but low 2D scaffold similarity to the template molecules. This ratio was 1.9 times higher than other state-of-the-art deep learning methods and rule- and virtual screening-based methods. Furthermore, we demonstrated that the model could generalize to new target proteins through fine-tuning with a small set of active compounds. Case studies have also shown the advantages and usefulness of DeepHop in practical scaffold hopping scenarios.Shuangjia ZhengZengrong LeiHaitao AiHongming ChenDaiguo DengYuedong YangBMCarticleDeep learningDrug designScaffold hoppingMolecular optimizationTransformer neural networkInformation technologyT58.5-58.64ChemistryQD1-999ENJournal of Cheminformatics, Vol 13, Iss 1, Pp 1-15 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Deep learning Drug design Scaffold hopping Molecular optimization Transformer neural network Information technology T58.5-58.64 Chemistry QD1-999
spellingShingle	Deep learning Drug design Scaffold hopping Molecular optimization Transformer neural network Information technology T58.5-58.64 Chemistry QD1-999 Shuangjia Zheng Zengrong Lei Haitao Ai Hongming Chen Daiguo Deng Yuedong Yang Deep scaffold hopping with multimodal transformer neural networks
description	Abstract Scaffold hopping is a central task of modern medicinal chemistry for rational drug design, which aims to design molecules of novel scaffolds sharing similar target biological activities toward known hit molecules. Traditionally, scaffolding hopping depends on searching databases of available compounds that can't exploit vast chemical space. In this study, we have re-formulated this task as a supervised molecule-to-molecule translation to generate hopped molecules novel in 2D structure but similar in 3D structure, as inspired by the fact that candidate compounds bind with their targets through 3D conformations. To efficiently train the model, we curated over 50 thousand pairs of molecules with increased bioactivity, similar 3D structure, but different 2D structure from public bioactivity database, which spanned 40 kinases commonly investigated by medicinal chemists. Moreover, we have designed a multimodal molecular transformer architecture by integrating molecular 3D conformer through a spatial graph neural network and protein sequence information through Transformer. The trained DeepHop model was shown able to generate around 70% molecules having improved bioactivity together with high 3D similarity but low 2D scaffold similarity to the template molecules. This ratio was 1.9 times higher than other state-of-the-art deep learning methods and rule- and virtual screening-based methods. Furthermore, we demonstrated that the model could generalize to new target proteins through fine-tuning with a small set of active compounds. Case studies have also shown the advantages and usefulness of DeepHop in practical scaffold hopping scenarios.
format	article
author	Shuangjia Zheng Zengrong Lei Haitao Ai Hongming Chen Daiguo Deng Yuedong Yang
author_facet	Shuangjia Zheng Zengrong Lei Haitao Ai Hongming Chen Daiguo Deng Yuedong Yang
author_sort	Shuangjia Zheng
title	Deep scaffold hopping with multimodal transformer neural networks
title_short	Deep scaffold hopping with multimodal transformer neural networks
title_full	Deep scaffold hopping with multimodal transformer neural networks
title_fullStr	Deep scaffold hopping with multimodal transformer neural networks
title_full_unstemmed	Deep scaffold hopping with multimodal transformer neural networks
title_sort	deep scaffold hopping with multimodal transformer neural networks
publisher	BMC
publishDate	2021
url	https://doaj.org/article/c9a4a58e594a43799741bcdab9961ad1
work_keys_str_mv	AT shuangjiazheng deepscaffoldhoppingwithmultimodaltransformerneuralnetworks AT zengronglei deepscaffoldhoppingwithmultimodaltransformerneuralnetworks AT haitaoai deepscaffoldhoppingwithmultimodaltransformerneuralnetworks AT hongmingchen deepscaffoldhoppingwithmultimodaltransformerneuralnetworks AT daiguodeng deepscaffoldhoppingwithmultimodaltransformerneuralnetworks AT yuedongyang deepscaffoldhoppingwithmultimodaltransformerneuralnetworks
_version_	1718429193682812928

Deep scaffold hopping with multimodal transformer neural networks

Ejemplares similares