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

Diverse clinical manifestations of porokeratosis caused by somatic variants and promoter hypermethylation of FDFT1.

The British journal of dermatology·2026
Same author

Defining a Chemical Space of π-Conjugated Hydrocarbons by Unit-Based Construction.

Journal of chemical information and modeling·2026
Same author

Rapidly Progressive Phaeohyphomycosis Caused by Pleurostoma richardsiae in an Immunosuppressed Patient.

The Journal of dermatology·2026
Same author

Advances in In silico predictive models for DDI prediction: Implications and practical applications in drug discovery.

Drug metabolism and pharmacokinetics·2026
Same author

Assessing the performance of multimodal large language models in experimental information extraction from liquid-liquid phase separation literature.

Scientific reports·2026
Same author

Elevated serum d-ROMs and lesional 4-HNE expression indicate oxidative stress in pemphigus.

Journal of dermatological science·2026

Related Experiment Video

Updated: Feb 9, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

2.6K

Machine learning accelerates MD-based binding pose prediction between ligands and proteins.

Kei Terayama1, Hiroaki Iwata2, Mitsugu Araki3

  • 1Department of Computational Biology and Medical Science, Graduate School of Frontier Sciences, The University of Tokyo, Chiba 277-8561, Japan.

Bioinformatics (Oxford, England)
|October 18, 2017
PubMed
Summary

This study introduces a machine learning approach, Best Arm Identification, to optimize molecular dynamics (MD) simulations for predicting protein-ligand binding poses. This method reduces computational cost without compromising accuracy in drug discovery.

More Related Videos

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

747
A Protocol for Computer-Based Protein Structure and Function Prediction
16:41

A Protocol for Computer-Based Protein Structure and Function Prediction

Published on: November 3, 2011

69.8K

Related Experiment Videos

Last Updated: Feb 9, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework
06:19

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

2.6K
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

747
A Protocol for Computer-Based Protein Structure and Function Prediction
16:41

A Protocol for Computer-Based Protein Structure and Function Prediction

Published on: November 3, 2011

69.8K

Area of Science:

  • Computational chemistry
  • Molecular modeling
  • Drug discovery

Background:

  • Accurate prediction of protein-ligand binding poses is crucial for structure-based drug design.
  • Current methods like MM-PBSA and MM-GBSA rely on Molecular Dynamics (MD) simulations, but accuracy depends on initial conditions.
  • Multiple MD runs with varying initial velocities can improve prediction accuracy.

Purpose of the Study:

  • To develop an optimized strategy for controlling the number of MD runs in protein-ligand binding pose prediction.
  • To reduce computational cost while maintaining or improving prediction accuracy.
  • To apply a machine learning method for efficient molecular simulations.

Main Methods:

  • Implementation of the Best Arm Identification (BAI) machine learning algorithm.
  • BAI is used to optimally control the number of MD simulation runs for each potential binding pose.
  • Experimental validation using three proteins and eight inhibitors.

Main Results:

  • The Best Arm Identification method effectively optimizes the number of MD runs required.
  • Substantial reduction in computational cost was achieved without sacrificing prediction accuracy.
  • The approach successfully identified correct binding poses with a minimum number of total simulation runs.

Conclusions:

  • The developed machine learning method offers an efficient way to perform molecular simulations.
  • This approach can be broadly applied to optimize various molecular simulations under computational constraints.
  • It significantly enhances the efficiency of structure-based drug design and binding free energy estimation.