Explore Protein Conformational Space With Variational Autoencoder
Molecular dynamics (MD) simulations have been actively used in the study of protein structure and function. However, extensive sampling in the protein conformational space requires large computational resources and takes a prohibitive amount of time. In this study, we demonstrated that variational a...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-11-01
|
| Series: | Frontiers in Molecular Biosciences |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmolb.2021.781635/full |
| _version_ | 1819003209977430016 |
|---|---|
| author | Hao Tian Xi Jiang Francesco Trozzi Sian Xiao Eric C. Larson Peng Tao |
| author_facet | Hao Tian Xi Jiang Francesco Trozzi Sian Xiao Eric C. Larson Peng Tao |
| author_sort | Hao Tian |
| collection | DOAJ |
| description | Molecular dynamics (MD) simulations have been actively used in the study of protein structure and function. However, extensive sampling in the protein conformational space requires large computational resources and takes a prohibitive amount of time. In this study, we demonstrated that variational autoencoders (VAEs), a type of deep learning model, can be employed to explore the conformational space of a protein through MD simulations. VAEs are shown to be superior to autoencoders (AEs) through a benchmark study, with low deviation between the training and decoded conformations. Moreover, we show that the learned latent space in the VAE can be used to generate unsampled protein conformations. Additional simulations starting from these generated conformations accelerated the sampling process and explored hidden spaces in the conformational landscape. |
| first_indexed | 2024-12-20T23:17:22Z |
| format | Article |
| id | doaj.art-0dd63e9213564b10acf17ca8968be3b0 |
| institution | Directory Open Access Journal |
| issn | 2296-889X |
| language | English |
| last_indexed | 2024-12-20T23:17:22Z |
| publishDate | 2021-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Molecular Biosciences |
| spelling | doaj.art-0dd63e9213564b10acf17ca8968be3b02022-12-21T19:23:36ZengFrontiers Media S.A.Frontiers in Molecular Biosciences2296-889X2021-11-01810.3389/fmolb.2021.781635781635Explore Protein Conformational Space With Variational AutoencoderHao Tian0Xi Jiang1Francesco Trozzi2Sian Xiao3Eric C. Larson4Peng Tao5Center for Research Computing, Center for Drug Discovery, Design, and Delivery (CD4), Department of Chemistry, Southern Methodist University, Dallas, TX, United StatesDepartment of Statistical Science, Southern Methodist University, Dallas, TX, United StatesCenter for Research Computing, Center for Drug Discovery, Design, and Delivery (CD4), Department of Chemistry, Southern Methodist University, Dallas, TX, United StatesCenter for Research Computing, Center for Drug Discovery, Design, and Delivery (CD4), Department of Chemistry, Southern Methodist University, Dallas, TX, United StatesDepartment of Computer Science, Southern Methodist University, Dallas, TX, United StatesCenter for Research Computing, Center for Drug Discovery, Design, and Delivery (CD4), Department of Chemistry, Southern Methodist University, Dallas, TX, United StatesMolecular dynamics (MD) simulations have been actively used in the study of protein structure and function. However, extensive sampling in the protein conformational space requires large computational resources and takes a prohibitive amount of time. In this study, we demonstrated that variational autoencoders (VAEs), a type of deep learning model, can be employed to explore the conformational space of a protein through MD simulations. VAEs are shown to be superior to autoencoders (AEs) through a benchmark study, with low deviation between the training and decoded conformations. Moreover, we show that the learned latent space in the VAE can be used to generate unsampled protein conformations. Additional simulations starting from these generated conformations accelerated the sampling process and explored hidden spaces in the conformational landscape.https://www.frontiersin.org/articles/10.3389/fmolb.2021.781635/fullprotein systemconformational spacevariational autoencodermolecular dynamicsdeep learning |
| spellingShingle | Hao Tian Xi Jiang Francesco Trozzi Sian Xiao Eric C. Larson Peng Tao Explore Protein Conformational Space With Variational Autoencoder Frontiers in Molecular Biosciences protein system conformational space variational autoencoder molecular dynamics deep learning |
| title | Explore Protein Conformational Space With Variational Autoencoder |
| title_full | Explore Protein Conformational Space With Variational Autoencoder |
| title_fullStr | Explore Protein Conformational Space With Variational Autoencoder |
| title_full_unstemmed | Explore Protein Conformational Space With Variational Autoencoder |
| title_short | Explore Protein Conformational Space With Variational Autoencoder |
| title_sort | explore protein conformational space with variational autoencoder |
| topic | protein system conformational space variational autoencoder molecular dynamics deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fmolb.2021.781635/full |
| work_keys_str_mv | AT haotian exploreproteinconformationalspacewithvariationalautoencoder AT xijiang exploreproteinconformationalspacewithvariationalautoencoder AT francescotrozzi exploreproteinconformationalspacewithvariationalautoencoder AT sianxiao exploreproteinconformationalspacewithvariationalautoencoder AT ericclarson exploreproteinconformationalspacewithvariationalautoencoder AT pengtao exploreproteinconformationalspacewithvariationalautoencoder |