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

Competition02:34

Competition

24.2K
When organisms require the same limited resources within an environment, they may have to compete for them. Competition is a net-negative interaction. Even if two competing individuals or populations do not interact directly, the overall fitness of both competitors is lowered as a result of not having full access to the limited resource.
24.2K

You might also read

Related Articles

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

Sort by
Same author

Gangrenous cholecystitis in a non-elderly man taking opioids.

Journal of rural medicine : JRM·2026
Same author

Biochemical and structural analyses of metallo-β-lactamase VIM-28: impact of substitutions at residues 224 and 228 on substrate profile, stability, and zinc affinity.

Microbiology spectrum·2026
Same author

Advanced measurement modeling of the Clock Drawing Test: A confirmatory factor analysis and item response theory-based validation of CDT-14.

Journal of Alzheimer's disease : JAD·2026
Same author

Structural Analysis of GodF, an O-glutamylation Enzyme Involved in Goadsporin Biosynthesis.

Journal of molecular biology·2026
Same author

Assessment of localized brain regions correlated with MMSE using VBM analysis of structural MRI in a Japanese sample.

Neuroimage. Reports·2025
Same author

Successful Tepotinib Treatment in <i>MET</i> Exon 14 Skipping Mutation-Positive Lung Adenocarcinoma With Idiopathic Pulmonary Fibrosis Through Nintedanib Combination.

Respirology case reports·2025

Related Experiment Video

Updated: Jan 2, 2026

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

782

Validation Study of QSAR/DNN Models Using the Competition Datasets.

Yoshiki Kato1, Shinji Hamada1, Hitoshi Goto1

  • 1Department of Computer Science and Engineering, Toyohashi University of Technology, 1-1 Hibarigaoka, Tempaku cho, Toyohashi, Aichi, 441-8580, Japan.

Molecular Informatics
|December 6, 2019
PubMed
Summary

Artificial intelligence (AI) and deep neural networks (DNNs) are revolutionizing molecular sciences. This study introduces a "Meister" setting for QSAR/DNN models, achieving champion-level performance by adjusting DNN parameters like mini-batch size.

Keywords:
Activity ChallengeChainer ChemistryDeep Neural NetworkMachine LearningMerck Molecular

More Related Videos

An R-Based Landscape Validation of a Competing Risk Model
05:37

An R-Based Landscape Validation of a Competing Risk Model

Published on: September 16, 2022

2.5K

Related Experiment Videos

Last Updated: Jan 2, 2026

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
06:19

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

782
An R-Based Landscape Validation of a Competing Risk Model
05:37

An R-Based Landscape Validation of a Competing Risk Model

Published on: September 16, 2022

2.5K

Area of Science:

  • Computational chemistry and cheminformatics.
  • Application of artificial intelligence in drug and material discovery.

Background:

  • Quantitative Structure-Activity Relationship (QSAR) models utilizing Deep Neural Networks (DNNs) have demonstrated superior predictive performance compared to traditional methods.
  • The integration of artificial intelligence (AI) with general-purpose artificial neural network (ANN) platforms and large chemical databases has spurred significant interest in molecular sciences.

Purpose of the Study:

  • To investigate various DNN settings for QSAR/DNN models to achieve predictive performance comparable to top competitors in the Kaggle QSAR competition.
  • To introduce an optimized DNN configuration, termed the "Meister" setting, for constructing high-performance QSAR/DNN models.

Main Methods:

  • Utilized a commonly available DNN model and trained numerous QSAR/DNN models.
  • Employed 15 datasets from the Kaggle QSAR competition to test various DNN configurations.
  • Compared performance against the champion team's results, focusing on R2 values.

Main Results:

  • Achieved R2 performance levels equivalent to the champion team using a general-purpose ANN platform.
  • Identified a key difference in DNN settings: reducing the mini-batch size improved performance.
  • Demonstrated that R2 performance is influenced by molecular activity type and the complexity of underlying biological or chemical processes.

Conclusions:

  • The "Meister" setting, particularly with adjusted mini-batch sizes, enables the creation of high-performance QSAR/DNN models.
  • This research validates the effectiveness of accessible DNN platforms and optimized settings for advancing drug and material discovery.
  • Understanding the relationship between molecular activity complexity and model performance is crucial for accurate predictions.