Cargando…

Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case

[Image: see text] Determining the redox potentials of protein cofactors and how they are influenced by their molecular neighborhoods is essential for basic research and many biotechnological applications, from biosensors and biocatalysis to bioremediation and bioelectronics. The laborious determinat...

Descripción completa

Detalles Bibliográficos
Autores principales: Galuzzi, Bruno Giovanni, Mirarchi, Antonio, Viganò, Edoardo Luca, De Gioia, Luca, Damiani, Chiara, Arrigoni, Federica
Formato: Online Artículo Texto
Lenguaje:English
Publicado: American Chemical Society 2022
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9554915/
https://www.ncbi.nlm.nih.gov/pubmed/36126254
http://dx.doi.org/10.1021/acs.jcim.2c00858
_version_ 1784806804552679424
author Galuzzi, Bruno Giovanni
Mirarchi, Antonio
Viganò, Edoardo Luca
De Gioia, Luca
Damiani, Chiara
Arrigoni, Federica
author_facet Galuzzi, Bruno Giovanni
Mirarchi, Antonio
Viganò, Edoardo Luca
De Gioia, Luca
Damiani, Chiara
Arrigoni, Federica
author_sort Galuzzi, Bruno Giovanni
collection PubMed
description [Image: see text] Determining the redox potentials of protein cofactors and how they are influenced by their molecular neighborhoods is essential for basic research and many biotechnological applications, from biosensors and biocatalysis to bioremediation and bioelectronics. The laborious determination of redox potential with current experimental technologies pushes forward the need for computational approaches that can reliably predict it. Although current computational approaches based on quantum and molecular mechanics are accurate, their large computational costs hinder their usage. In this work, we explored the possibility of using more efficient QSPR models based on machine learning (ML) for the prediction of protein redox potential, as an alternative to classical approaches. As a proof of concept, we focused on flavoproteins, one of the most important families of enzymes directly involved in redox processes. To train and test different ML models, we retrieved a dataset of flavoproteins with a known midpoint redox potential (E(m)) and 3D structure. The features of interest, accounting for both short- and long-range effects of the protein matrix on the flavin cofactor, have been automatically extracted from each protein PDB file. Our best ML model (XGB) has a performance error below 1 kcal/mol (∼36 mV), comparing favorably to more sophisticated computational approaches. We also provided indications on the features that mostly affect the E(m) value, and when possible, we rationalized them on the basis of previous studies.
format Online
Article
Text
id pubmed-9554915
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher American Chemical Society
record_format MEDLINE/PubMed
spelling pubmed-95549152022-10-13 Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case Galuzzi, Bruno Giovanni Mirarchi, Antonio Viganò, Edoardo Luca De Gioia, Luca Damiani, Chiara Arrigoni, Federica J Chem Inf Model [Image: see text] Determining the redox potentials of protein cofactors and how they are influenced by their molecular neighborhoods is essential for basic research and many biotechnological applications, from biosensors and biocatalysis to bioremediation and bioelectronics. The laborious determination of redox potential with current experimental technologies pushes forward the need for computational approaches that can reliably predict it. Although current computational approaches based on quantum and molecular mechanics are accurate, their large computational costs hinder their usage. In this work, we explored the possibility of using more efficient QSPR models based on machine learning (ML) for the prediction of protein redox potential, as an alternative to classical approaches. As a proof of concept, we focused on flavoproteins, one of the most important families of enzymes directly involved in redox processes. To train and test different ML models, we retrieved a dataset of flavoproteins with a known midpoint redox potential (E(m)) and 3D structure. The features of interest, accounting for both short- and long-range effects of the protein matrix on the flavin cofactor, have been automatically extracted from each protein PDB file. Our best ML model (XGB) has a performance error below 1 kcal/mol (∼36 mV), comparing favorably to more sophisticated computational approaches. We also provided indications on the features that mostly affect the E(m) value, and when possible, we rationalized them on the basis of previous studies. American Chemical Society 2022-09-20 2022-10-10 /pmc/articles/PMC9554915/ /pubmed/36126254 http://dx.doi.org/10.1021/acs.jcim.2c00858 Text en © 2022 The Authors. Published by American Chemical Society https://creativecommons.org/licenses/by/4.0/Permits the broadest form of re-use including for commercial purposes, provided that author attribution and integrity are maintained (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Galuzzi, Bruno Giovanni
Mirarchi, Antonio
Viganò, Edoardo Luca
De Gioia, Luca
Damiani, Chiara
Arrigoni, Federica
Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title_full Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title_fullStr Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title_full_unstemmed Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title_short Machine Learning for Efficient Prediction of Protein Redox Potential: The Flavoproteins Case
title_sort machine learning for efficient prediction of protein redox potential: the flavoproteins case
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9554915/
https://www.ncbi.nlm.nih.gov/pubmed/36126254
http://dx.doi.org/10.1021/acs.jcim.2c00858
work_keys_str_mv AT galuzzibrunogiovanni machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase
AT mirarchiantonio machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase
AT viganoedoardoluca machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase
AT degioialuca machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase
AT damianichiara machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase
AT arrigonifederica machinelearningforefficientpredictionofproteinredoxpotentialtheflavoproteinscase