Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Homotopy reservoir computing: Harnessing chaos for computation.

Chaos (Woodbury, N.Y.)·2025
Same author

Signal-noise separation using unsupervised reservoir computing.

Chaos (Woodbury, N.Y.)·2025
Same author

Pd-Modified Microneedle Array Sensor Integration with Deep Learning for Predicting Silica Aerogel Properties in Real Time.

ACS applied materials & interfaces·2025
Same author

Periodically driven thermodynamic systems under vanishingly small viscous drives.

Physical review. E·2025
Same author

Oscillating states of driven Langevin systems under large viscous drives.

Physical review. E·2023
Same author

Oscillating states of periodically driven anharmonic Langevin systems.

Physical review. E·2021

Related Experiment Video

Updated: May 1, 2026

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences
06:49

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Published on: June 16, 2014

16.5K

Stochastic intracellular calcium dynamics show preserved structures identified by deep learning classification.

Jaesung Choi1, Athokpam Langlen Chanu2,3, Shakul Awasthi4

  • 1Center for Artificial Intelligence and Natural Sciences, Korea Institute for Advanced Study (KIAS), Seoul, Republic of Korea.

Plos Computational Biology
|April 29, 2026
PubMed
Summary
This summary is machine-generated.

Deterministic cellular dynamics persist despite biological noise. A large-kernel convolutional neural network (LKCNN) accurately classifies cell signaling states, even with significant fluctuations, revealing robust organizational principles.

More Related Videos

Functional Calcium Imaging in Developing Cortical Networks
16:33

Functional Calcium Imaging in Developing Cortical Networks

Published on: October 22, 2011

41.3K
Imaging Local Ca2+ Signals in Cultured Mammalian Cells
09:30

Imaging Local Ca2+ Signals in Cultured Mammalian Cells

Published on: March 3, 2015

10.3K

Related Experiment Videos

Last Updated: May 1, 2026

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences
06:49

Automated Analysis of Dynamic Ca2+ Signals in Image Sequences

Published on: June 16, 2014

16.5K
Functional Calcium Imaging in Developing Cortical Networks
16:33

Functional Calcium Imaging in Developing Cortical Networks

Published on: October 22, 2011

41.3K
Imaging Local Ca2+ Signals in Cultured Mammalian Cells
09:30

Imaging Local Ca2+ Signals in Cultured Mammalian Cells

Published on: March 3, 2015

10.3K

Area of Science:

  • Cellular dynamics and signaling
  • Computational biology and machine learning
  • Biophysics of stochastic systems

Background:

  • Intracellular calcium ions (Ca2+) exhibit complex dynamics crucial for cellular functions, health, and disease.
  • Deterministic models predict distinct dynamical regimes (e.g., oscillations, chaos), but biological systems are inherently stochastic due to finite molecular numbers.
  • Conventional statistical methods struggle to distinguish these dynamical states under realistic noise levels.

Purpose of the Study:

  • To investigate whether parameter-dependent organizational principles in cellular dynamics persist under realistic biological noise.
  • To develop and validate a computational framework capable of detecting these principles despite noise-induced indistinguishability of dynamical states.

Main Methods:

  • Utilized chemical Langevin equations to generate synthetic training data reflecting realistic intrinsic fluctuations.
  • Developed a large-kernel convolutional neural network (LKCNN) to capture global dynamical features across varying noise levels.
  • Validated the LKCNN using experimental Ca2+ data from pancreatic beta-cells and other cell lines (WT-HEK293, STIM-KO, ORAI TKO).

Main Results:

  • The LKCNN achieved ~90% accuracy in classifying eight distinct dynamical states from synthetic data, even with high noise levels.
  • On experimental Ca2+ data, the LKCNN achieved 96.8% accuracy, significantly outperforming conventional methods like Support Vector Machine (54.0%) and Random Forest (51.6%).
  • Demonstrated that deterministic organizational signatures remain detectable despite biological noise.

Conclusions:

  • Parameter-dependent dynamical structures represent robust principles governing cellular function, persisting through biological noise.
  • Sophisticated pattern recognition, like LKCNN, can bridge theoretical deterministic dynamics and noisy biological reality.
  • This framework offers a novel approach for extracting meaningful dynamical information from inherently stochastic oscillatory biological processes.