BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network
Background: For decades, the rate of solving new biomolecular structures has been exceeding that at which their manual classification and feature characterisation can be carried out efficiently. Therefore, a new comprehensive and holistic tool for their examination is needed. Methods: Here we propos...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-03-01
|
Series: | International Journal of Molecular Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/1422-0067/23/6/2966 |
_version_ | 1797471006777933824 |
---|---|
author | Albert Roethel Piotr Biliński Takao Ishikawa |
author_facet | Albert Roethel Piotr Biliński Takao Ishikawa |
author_sort | Albert Roethel |
collection | DOAJ |
description | Background: For decades, the rate of solving new biomolecular structures has been exceeding that at which their manual classification and feature characterisation can be carried out efficiently. Therefore, a new comprehensive and holistic tool for their examination is needed. Methods: Here we propose the Biological Sequence and Structure Network (BioS2Net), which is a novel deep neural network architecture that extracts both sequential and structural information of biomolecules. Our architecture consists of four main parts: (i) a sequence convolutional extractor, (ii) a 3D structure extractor, (iii) a 3D structure-aware sequence temporal network, as well as (iv) a fusion and classification network. Results: We have evaluated our approach using two protein fold classification datasets. BioS2Net achieved a 95.4% mean class accuracy on the eDD dataset and a 76% mean class accuracy on the F184 dataset. The accuracy of BioS2Net obtained on the eDD dataset was comparable to results achieved by previously published methods, confirming that the algorithm described in this article is a top-class solution for protein fold recognition. Conclusions: BioS2Net is a novel tool for the holistic examination of biomolecules of known structure and sequence. It is a reliable tool for protein analysis and their unified representation as feature vectors. |
first_indexed | 2024-03-09T19:43:26Z |
format | Article |
id | doaj.art-7f31b829f57d43bda3b5f8113a3010bf |
institution | Directory Open Access Journal |
issn | 1661-6596 1422-0067 |
language | English |
last_indexed | 2024-03-09T19:43:26Z |
publishDate | 2022-03-01 |
publisher | MDPI AG |
record_format | Article |
series | International Journal of Molecular Sciences |
spelling | doaj.art-7f31b829f57d43bda3b5f8113a3010bf2023-11-24T01:29:51ZengMDPI AGInternational Journal of Molecular Sciences1661-65961422-00672022-03-01236296610.3390/ijms23062966BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural NetworkAlbert Roethel0Piotr Biliński1Takao Ishikawa2Department of Molecular Biology, Institute of Biochemistry, Faculty of Biology, University of Warsaw, 02-096 Warsaw, PolandInstitute of Informatics, Faculty of Mathematics, Informatics and Mechanics, University of Warsaw, 02-097 Warsaw, PolandDepartment of Molecular Biology, Institute of Biochemistry, Faculty of Biology, University of Warsaw, 02-096 Warsaw, PolandBackground: For decades, the rate of solving new biomolecular structures has been exceeding that at which their manual classification and feature characterisation can be carried out efficiently. Therefore, a new comprehensive and holistic tool for their examination is needed. Methods: Here we propose the Biological Sequence and Structure Network (BioS2Net), which is a novel deep neural network architecture that extracts both sequential and structural information of biomolecules. Our architecture consists of four main parts: (i) a sequence convolutional extractor, (ii) a 3D structure extractor, (iii) a 3D structure-aware sequence temporal network, as well as (iv) a fusion and classification network. Results: We have evaluated our approach using two protein fold classification datasets. BioS2Net achieved a 95.4% mean class accuracy on the eDD dataset and a 76% mean class accuracy on the F184 dataset. The accuracy of BioS2Net obtained on the eDD dataset was comparable to results achieved by previously published methods, confirming that the algorithm described in this article is a top-class solution for protein fold recognition. Conclusions: BioS2Net is a novel tool for the holistic examination of biomolecules of known structure and sequence. It is a reliable tool for protein analysis and their unified representation as feature vectors.https://www.mdpi.com/1422-0067/23/6/2966deep neural networkfeature vectorproteinprotein fold classification |
spellingShingle | Albert Roethel Piotr Biliński Takao Ishikawa BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network International Journal of Molecular Sciences deep neural network feature vector protein protein fold classification |
title | BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network |
title_full | BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network |
title_fullStr | BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network |
title_full_unstemmed | BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network |
title_short | BioS2Net: Holistic Structural and Sequential Analysis of Biomolecules Using a Deep Neural Network |
title_sort | bios2net holistic structural and sequential analysis of biomolecules using a deep neural network |
topic | deep neural network feature vector protein protein fold classification |
url | https://www.mdpi.com/1422-0067/23/6/2966 |
work_keys_str_mv | AT albertroethel bios2netholisticstructuralandsequentialanalysisofbiomoleculesusingadeepneuralnetwork AT piotrbilinski bios2netholisticstructuralandsequentialanalysisofbiomoleculesusingadeepneuralnetwork AT takaoishikawa bios2netholisticstructuralandsequentialanalysisofbiomoleculesusingadeepneuralnetwork |