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 Experiment Videos

A method for quantifying and visualizing the diversity of QSAR models.

Sergei Izrailev1, Dimitris K Agrafiotis

  • 13-Dimensional Pharmaceuticals, Inc., 8 Clarke Drive, Cranbury, NJ 08512, USA. sergei.izrailev@3dp.com

Journal of Molecular Graphics & Modelling
|June 5, 2004
PubMed
Summary

Feature selection for quantitative structure-activity relationships (QSAR) can yield similar models using different features. We introduce a similarity measure to identify these models, aiding QSAR analysis.

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

Epilogue to the Gerald Maggiora Festschrift: a tribute to an exemplary mentor, colleague, collaborator, and innovator.

Journal of computer-aided molecular design·2022
Same author

Accurate prediction of clinical stroke scales and improved biomarkers of motor impairment from robotic measurements.

PloS one·2021
Same author

Xcellerate Investigator Portal: A New Web-Based Tool for Online Delivery of Central Laboratory Data, Reports, and Communications to Clinical Sites.

SLAS technology·2020
Same author

A new risk and issue management system to improve productivity, quality, and compliance in clinical trials.

JAMIA open·2020
Same author

A dimensional warehouse for integrating operational data from clinical trials.

Database : the journal of biological databases and curation·2019
Same author

A late-binding, distributed, NoSQL warehouse for integrating patient data from clinical trials.

Database : the journal of biological databases and curation·2019

Area of Science:

  • Computational chemistry
  • Medicinal chemistry
  • Bioinformatics

Background:

  • Feature selection is crucial for developing predictive quantitative structure-activity relationships (QSAR) models.
  • Stochastic feature selection algorithms can produce varied results due to initialization.
  • Identifying QSAR models that capture similar information despite using different features is challenging.

Purpose of the Study:

  • To introduce a novel similarity measure for quantitative structure-activity relationship (QSAR) models.
  • To address the challenge of recognizing models with highly correlated underlying features.
  • To facilitate the analysis and visualization of QSAR model spaces.

Main Methods:

  • Development of a new similarity metric for QSAR models.

Related Experiment Videos

  • Utilizing stochastic proximity embedding (SPE) or multi-dimensional scaling (MDS) for visualization.
  • Application of the similarity measure to analyze feature selection results.

    • Main Results:

    • A new measure effectively quantifies similarity between QSAR models based on feature correlation.
    • Visualizations using SPE or MDS reveal relationships within the QSAR model space.
    • The method aids in distinguishing genuinely different models from those with correlated features.

    Conclusions:

    • The proposed similarity measure enhances the analysis of QSAR models derived from feature selection.
    • Visualizing model similarity aids in understanding and selecting robust QSAR models.
    • This approach improves the interpretability and reliability of QSAR studies.