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

VSMP: a novel variable selection and modeling method based on the prediction.

Shu-Shen Liu1, Hai-Ling Liu, Chun-Sheng Yin

  • 1State Key Laboratory of Pollution Control and Resources Reuse, Department of Environmental Science & Engineering, Nanjing University, Nanjing 210093, P. R. China. ssliuhl@263.net

Journal of Chemical Information and Computer Sciences
|May 28, 2003
PubMed
Summary
This summary is machine-generated.

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

Time-dependent shift from antagonism to synergism in mixture toxicity mediated by GFP-modified Caenorhabditis elegans.

Ecotoxicology and environmental safety·2026
Same author

CHK1 activates mitophagy to attenuate cardiac aging via inhibiting AHSA1-ubiquitination.

Redox biology·2026
Same author

Interpretable multi-species QSAR modeling for ecotoxicological hazard prediction of aquatic pollutants.

Environment international·2026
Same author

An Investigation of Iron-Induced Reconstruction of a Molecular Cobalt Precatalyst for Efficient Oxygen Evolution.

ACS applied materials & interfaces·2026
Same author

Competitive effect-derived surface-enhanced Raman scattering for analysis of glutathione based on a gold nanotriangle-modified nanocapillary.

Analytica chimica acta·2025
Same author

Chronicle of mixture effects: Sensitivity differences in multiple toxicity endpoints of transgenic C. elegans and assessment of combined toxicity interactions.

Toxicology letters·2025
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
See all related articles

Choosing optimal molecular descriptors for quantitative structure-activity relationship (QSAR) studies is challenging. A novel variable selection and modeling method (VSMP) improves QSAR by using interrelation (r(int)) and cross-validation (q(2)) coefficients for better descriptor selection.

Area of Science:

  • Medicinal Chemistry
  • Computational Chemistry
  • Cheminformatics

Background:

  • Quantitative Structure-Activity Relationship (QSAR) studies increasingly utilize molecular descriptors.
  • Selecting appropriate descriptors is crucial but lacks standardized rules.
  • Existing variable selection methods include stepwise, PLS/PCA, neural networks, and genetic algorithms.

Purpose of the Study:

  • To develop a novel variable selection and modeling method (VSMP) for QSAR.
  • To enhance descriptor selection by incorporating interrelation (r(int)) and cross-validation (q(2)) statistics into All-Subsets Regression (ASR).

Main Methods:

  • Developed a new method, VSMP, integrating r(int) and q(2) statistics into ASR.
  • Utilized leave-one-out (LOO) cross-validation to guide subset selection via q(2) or RMSEP.

Related Experiment Videos

  • Employed r(int) to accelerate the search for optimal descriptor subsets.

    • Main Results:

    • The VSMP method demonstrated good performance on the Selwood dataset (31 compounds, 53 descriptors).
    • VSMP's selection process is controlled by cross-validation statistics (q(2)/RMSEP), not just modeling R(2).
    • The inclusion of r(int) improved the speed of identifying optimal subsets.

    Conclusions:

    • VSMP offers an effective approach for selecting molecular descriptors in QSAR.
    • The method improves upon traditional ASR by optimizing subset selection using cross-validation metrics.
    • VSMP provides a faster and more robust method for QSAR modeling.