deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction

Coarse-grained models have provided researchers with greatly improved computational efficiency in modeling structures and dynamics of biomacromolecules, but, to be practically useful, they need fast and accurate conversion methods back to the all-atom representation. Reconstruction of atomic details...

Full description

Saved in:
Bibliographic Details
Main Authors: Justyna D. Kryś, Maksymilian Głowacki , Piotr Śmieja , Dominik Gront
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Biomolecules
Subjects:
Online Access:https://www.mdpi.com/2218-273X/14/11/1448
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850217529565249536
author Justyna D. Kryś
Maksymilian Głowacki 
Piotr Śmieja 
Dominik Gront
author_facet Justyna D. Kryś
Maksymilian Głowacki 
Piotr Śmieja 
Dominik Gront
author_sort Justyna D. Kryś
collection DOAJ
description Coarse-grained models have provided researchers with greatly improved computational efficiency in modeling structures and dynamics of biomacromolecules, but, to be practically useful, they need fast and accurate conversion methods back to the all-atom representation. Reconstruction of atomic details may also be required in the case of some experimental methods, like electron microscopy, which may provide C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula>-only structures. In this contribution, we present a new method for recovery of all backbone atom positions from just the C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> coordinates. Our approach, called deepBBQ, uses a deep convolutional neural network to predict a single internal coordinate per peptide plate, based on C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> trace geometric features, and then proceeds to recalculate the cartesian coordinates based on the assumption that the peptide plate atoms lie in the same plane. Extensive comparison with similar programs shows that our solution is accurate and cost-efficient. The deepBBQ program is available as part of the open-source bioinformatics toolkit Bioshell and is free for download and the documentation is available online.
format Article
id doaj-art-18adcceb0b8943b690b3e95db110e454
institution OA Journals
issn 2218-273X
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Biomolecules
spelling doaj-art-18adcceb0b8943b690b3e95db110e4542025-08-20T02:08:02ZengMDPI AGBiomolecules2218-273X2024-11-011411144810.3390/biom14111448deepBBQ: A Deep Learning Approach to the Protein Backbone ReconstructionJustyna D. Kryś0Maksymilian Głowacki 1Piotr Śmieja 2Dominik Gront3Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, PolandFaculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, PolandFaculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, PolandFaculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, PolandCoarse-grained models have provided researchers with greatly improved computational efficiency in modeling structures and dynamics of biomacromolecules, but, to be practically useful, they need fast and accurate conversion methods back to the all-atom representation. Reconstruction of atomic details may also be required in the case of some experimental methods, like electron microscopy, which may provide C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula>-only structures. In this contribution, we present a new method for recovery of all backbone atom positions from just the C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> coordinates. Our approach, called deepBBQ, uses a deep convolutional neural network to predict a single internal coordinate per peptide plate, based on C<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> trace geometric features, and then proceeds to recalculate the cartesian coordinates based on the assumption that the peptide plate atoms lie in the same plane. Extensive comparison with similar programs shows that our solution is accurate and cost-efficient. The deepBBQ program is available as part of the open-source bioinformatics toolkit Bioshell and is free for download and the documentation is available online.https://www.mdpi.com/2218-273X/14/11/1448atomic reconstructionstructural bioinformaticsdeep learningcoarse-grained modelingconvolutional neural network
spellingShingle Justyna D. Kryś
Maksymilian Głowacki 
Piotr Śmieja 
Dominik Gront
deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
Biomolecules
atomic reconstruction
structural bioinformatics
deep learning
coarse-grained modeling
convolutional neural network
title deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
title_full deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
title_fullStr deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
title_full_unstemmed deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
title_short deepBBQ: A Deep Learning Approach to the Protein Backbone Reconstruction
title_sort deepbbq a deep learning approach to the protein backbone reconstruction
topic atomic reconstruction
structural bioinformatics
deep learning
coarse-grained modeling
convolutional neural network
url https://www.mdpi.com/2218-273X/14/11/1448
work_keys_str_mv AT justynadkrys deepbbqadeeplearningapproachtotheproteinbackbonereconstruction
AT maksymiliangłowacki deepbbqadeeplearningapproachtotheproteinbackbonereconstruction
AT piotrsmieja deepbbqadeeplearningapproachtotheproteinbackbonereconstruction
AT dominikgront deepbbqadeeplearningapproachtotheproteinbackbonereconstruction