A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor

Autores
Buena Maizon, Héctor; Barrantes, Francisco José
Año de publicación
2022
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Fil: Buena Maizon, Héctor. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; Argentina
Fil: Buena Maizon, Héctor. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Barrantes, Francisco José. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Barrantes, Francisco José. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; Argentina
Abstract: We present a concatenated deep-learning multiple neural network system for the analysis of single-molecule trajectories. We apply this machine learning-based analysis to characterize the translational diffusion of the nicotinic acetylcholine receptor at the plasma membrane, experimentally interrogated using superresolution optical microscopy. The receptor protein displays a heterogeneous diffusion behavior that goes beyond the ensemble level, with individual trajectories exhibiting more than one diffusive state, requiring the optimization of the neural networks through a hyperparameter analysis for different numbers of steps and durations, especially for short trajectories (<50 steps) where the accuracy of the models is most sensitive to localization errors. We next use the statistical models to test for Brownian, continuous-time random walk and fractional Brownian motion, and introduce and implement an additional, two-state model combining Brownian walks and obstructed diffusion mechanisms, enabling us to partition the two-state trajectories into segments, each of which is independently subjected to multiple analysis. The concatenated multi-network system evaluates and selects those physical models that most accurately describe the receptor’s translational diffusion. We show that the two-state Brownian-obstructed diffusion model can account for the experimentally observed anomalous diffusion (mostly subdiffusive) of the population and the heterogeneous single-molecule behavior, accurately describing the majority (72.5 to 88.7% for α-bungarotoxin-labeled receptor and between 73.5 and 90.3% for antibody-labeled molecules) of the experimentally observed trajectories, with only ~15% of the trajectories fitting to the fractional Brownian motion model.
Fuente
Briefings in Bioinformatics Vol.23, No.1, 2022
Materia
INTELIGENCIA ARTIFICIAL
APRENDIZAJE AUTOMÁTICO
APRENDIZAJE PROFUNDO
PROTEÍNA DE MEMBRANA
RECEPTOR DE NEUROTRANSMISORES
RECEPTOR DE ACETILCOLINA
COLESTEROL
SEGUIMIENTO DE PARTÍCULAS INDIVIDUALES
MICROSCOPÍA DE SUPERRESOLUCIÓN
Nivel de accesibilidad
acceso embargado
Condiciones de uso
Repositorio
Repositorio Institucional (UCA)
Institución
Pontificia Universidad Católica Argentina
OAI Identificador
oai:ucacris:123456789/14114

id RIUCA_dfdf4f1236314aff8216175003635f73
oai_identifier_str oai:ucacris:123456789/14114
network_acronym_str RIUCA
repository_id_str 2585
network_name_str Repositorio Institucional (UCA)
spelling A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptorBuena Maizon, HéctorBarrantes, Francisco JoséINTELIGENCIA ARTIFICIALAPRENDIZAJE AUTOMÁTICOAPRENDIZAJE PROFUNDOPROTEÍNA DE MEMBRANARECEPTOR DE NEUROTRANSMISORESRECEPTOR DE ACETILCOLINACOLESTEROLSEGUIMIENTO DE PARTÍCULAS INDIVIDUALESMICROSCOPÍA DE SUPERRESOLUCIÓNFil: Buena Maizon, Héctor. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; ArgentinaFil: Buena Maizon, Héctor. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Barrantes, Francisco José. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Barrantes, Francisco José. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; ArgentinaAbstract: We present a concatenated deep-learning multiple neural network system for the analysis of single-molecule trajectories. We apply this machine learning-based analysis to characterize the translational diffusion of the nicotinic acetylcholine receptor at the plasma membrane, experimentally interrogated using superresolution optical microscopy. The receptor protein displays a heterogeneous diffusion behavior that goes beyond the ensemble level, with individual trajectories exhibiting more than one diffusive state, requiring the optimization of the neural networks through a hyperparameter analysis for different numbers of steps and durations, especially for short trajectories (<50 steps) where the accuracy of the models is most sensitive to localization errors. We next use the statistical models to test for Brownian, continuous-time random walk and fractional Brownian motion, and introduce and implement an additional, two-state model combining Brownian walks and obstructed diffusion mechanisms, enabling us to partition the two-state trajectories into segments, each of which is independently subjected to multiple analysis. The concatenated multi-network system evaluates and selects those physical models that most accurately describe the receptor’s translational diffusion. We show that the two-state Brownian-obstructed diffusion model can account for the experimentally observed anomalous diffusion (mostly subdiffusive) of the population and the heterogeneous single-molecule behavior, accurately describing the majority (72.5 to 88.7% for α-bungarotoxin-labeled receptor and between 73.5 and 90.3% for antibody-labeled molecules) of the experimentally observed trajectories, with only ~15% of the trajectories fitting to the fractional Brownian motion model.Oxford University Pressinfo:eu-repo/date/embargoEnd/2100-01-012022info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfhttps://repositorio.uca.edu.ar/handle/123456789/141141477-4054 (online)10.1093/bib/bbab435Buena Maizon, H., Barrantes, F. J. A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor [en línea]. Briefings in Bioinformatics. 2022, 23 (1). doi: https://doi.org/10.1093/bib/bbab435. Disponible en: https://repositorio.uca.edu.ar/handle/123456789/14114Briefings in Bioinformatics Vol.23, No.1, 2022reponame:Repositorio Institucional (UCA)instname:Pontificia Universidad Católica Argentinaenginfo:eu-repo/semantics/embargoedAccess2025-07-03T10:58:36Zoai:ucacris:123456789/14114instacron:UCAInstitucionalhttps://repositorio.uca.edu.ar/Universidad privadaNo correspondehttps://repositorio.uca.edu.ar/oaiclaudia_fernandez@uca.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:25852025-07-03 10:58:36.739Repositorio Institucional (UCA) - Pontificia Universidad Católica Argentinafalse
dc.title.none.fl_str_mv A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
title A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
spellingShingle A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
Buena Maizon, Héctor
INTELIGENCIA ARTIFICIAL
APRENDIZAJE AUTOMÁTICO
APRENDIZAJE PROFUNDO
PROTEÍNA DE MEMBRANA
RECEPTOR DE NEUROTRANSMISORES
RECEPTOR DE ACETILCOLINA
COLESTEROL
SEGUIMIENTO DE PARTÍCULAS INDIVIDUALES
MICROSCOPÍA DE SUPERRESOLUCIÓN
title_short A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
title_full A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
title_fullStr A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
title_full_unstemmed A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
title_sort A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor
dc.creator.none.fl_str_mv Buena Maizon, Héctor
Barrantes, Francisco José
author Buena Maizon, Héctor
author_facet Buena Maizon, Héctor
Barrantes, Francisco José
author_role author
author2 Barrantes, Francisco José
author2_role author
dc.subject.none.fl_str_mv INTELIGENCIA ARTIFICIAL
APRENDIZAJE AUTOMÁTICO
APRENDIZAJE PROFUNDO
PROTEÍNA DE MEMBRANA
RECEPTOR DE NEUROTRANSMISORES
RECEPTOR DE ACETILCOLINA
COLESTEROL
SEGUIMIENTO DE PARTÍCULAS INDIVIDUALES
MICROSCOPÍA DE SUPERRESOLUCIÓN
topic INTELIGENCIA ARTIFICIAL
APRENDIZAJE AUTOMÁTICO
APRENDIZAJE PROFUNDO
PROTEÍNA DE MEMBRANA
RECEPTOR DE NEUROTRANSMISORES
RECEPTOR DE ACETILCOLINA
COLESTEROL
SEGUIMIENTO DE PARTÍCULAS INDIVIDUALES
MICROSCOPÍA DE SUPERRESOLUCIÓN
dc.description.none.fl_txt_mv Fil: Buena Maizon, Héctor. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; Argentina
Fil: Buena Maizon, Héctor. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Barrantes, Francisco José. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Barrantes, Francisco José. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; Argentina
Abstract: We present a concatenated deep-learning multiple neural network system for the analysis of single-molecule trajectories. We apply this machine learning-based analysis to characterize the translational diffusion of the nicotinic acetylcholine receptor at the plasma membrane, experimentally interrogated using superresolution optical microscopy. The receptor protein displays a heterogeneous diffusion behavior that goes beyond the ensemble level, with individual trajectories exhibiting more than one diffusive state, requiring the optimization of the neural networks through a hyperparameter analysis for different numbers of steps and durations, especially for short trajectories (<50 steps) where the accuracy of the models is most sensitive to localization errors. We next use the statistical models to test for Brownian, continuous-time random walk and fractional Brownian motion, and introduce and implement an additional, two-state model combining Brownian walks and obstructed diffusion mechanisms, enabling us to partition the two-state trajectories into segments, each of which is independently subjected to multiple analysis. The concatenated multi-network system evaluates and selects those physical models that most accurately describe the receptor’s translational diffusion. We show that the two-state Brownian-obstructed diffusion model can account for the experimentally observed anomalous diffusion (mostly subdiffusive) of the population and the heterogeneous single-molecule behavior, accurately describing the majority (72.5 to 88.7% for α-bungarotoxin-labeled receptor and between 73.5 and 90.3% for antibody-labeled molecules) of the experimentally observed trajectories, with only ~15% of the trajectories fitting to the fractional Brownian motion model.
description Fil: Buena Maizon, Héctor. Pontificia Universidad Católica Argentina. Instituto de Investigaciones Biomédicas. Laboratorio de Neurobiología Molecular; Argentina
publishDate 2022
dc.date.none.fl_str_mv 2022
info:eu-repo/date/embargoEnd/2100-01-01
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://repositorio.uca.edu.ar/handle/123456789/14114
1477-4054 (online)
10.1093/bib/bbab435
Buena Maizon, H., Barrantes, F. J. A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor [en línea]. Briefings in Bioinformatics. 2022, 23 (1). doi: https://doi.org/10.1093/bib/bbab435. Disponible en: https://repositorio.uca.edu.ar/handle/123456789/14114
url https://repositorio.uca.edu.ar/handle/123456789/14114
identifier_str_mv 1477-4054 (online)
10.1093/bib/bbab435
Buena Maizon, H., Barrantes, F. J. A deep learning-based approach to model anomalous diffusion of membrane proteins: the case of the nicotinic acetylcholine receptor [en línea]. Briefings in Bioinformatics. 2022, 23 (1). doi: https://doi.org/10.1093/bib/bbab435. Disponible en: https://repositorio.uca.edu.ar/handle/123456789/14114
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/embargoedAccess
eu_rights_str_mv embargoedAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Oxford University Press
publisher.none.fl_str_mv Oxford University Press
dc.source.none.fl_str_mv Briefings in Bioinformatics Vol.23, No.1, 2022
reponame:Repositorio Institucional (UCA)
instname:Pontificia Universidad Católica Argentina
reponame_str Repositorio Institucional (UCA)
collection Repositorio Institucional (UCA)
instname_str Pontificia Universidad Católica Argentina
repository.name.fl_str_mv Repositorio Institucional (UCA) - Pontificia Universidad Católica Argentina
repository.mail.fl_str_mv claudia_fernandez@uca.edu.ar
_version_ 1836638362262306816
score 13.13397