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

Correlation of graph-theoretical parameters with biological activity

G G Cash1, J J Breen

  • 1United States Environmental Protection Agency, Office of Pollution Prevention and Toxics, Washington, D.C. 20460.

Journal of Chemical Information and Computer Sciences
|March 1, 1993
PubMed
Summary

Principal component analysis (PCA) successfully correlated graph-theoretical indices with biological activity in substituted isonicotinic hydrazides, outperforming regression models. Removing outliers further improved predictive accuracy for drug discovery.

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

Peer reviewed: promoting green chemistry initiatives.

Environmental science & technology·2011
Same author

Direct imaging of domains in the Lbeta' state of 1,2-dipalmitoylphosphatidylcholine bilayers.

Physical review. E, Statistical, nonlinear, and soft matter physics·2005
Same author

Polynomial expressions for the hyper-Wiener index of extended hydrocarbon networks.

Computers & chemistry·2002
Same author

Emerging technologies for large-scale screening of human tissues and fluids in the study of severe psychiatric disease.

The international journal of neuropsychopharmacology·2001
Same author

Prediction of the genotoxicity of aromatic and heteroaromatic amines using electrotopological state indices.

Mutation research·2001
Same author

The permanental polynomial.

Journal of chemical information and computer sciences·2000

Area of Science:

  • * Cheminformatics and Quantitative Structure-Activity Relationships (QSAR).
  • * Computational chemistry and drug design.

Background:

  • * Previous studies by Ośmialowski and Kaliszan explored correlations between graph-theoretical indices and biological activity of substituted isonicotinic hydrazides using regression, but were unsuccessful.
  • * Simple and multiple regression failed to establish significant correlations in the existing dataset.

Purpose of the Study:

  • * To re-evaluate the dataset using advanced multivariate statistical methods to find correlations with biological activity.
  • * To compare the predictive power of principal component analysis (PCA) against traditional regression techniques.

Main Methods:

  • * Application of Principal Component Analysis (PCA) to the dataset of graph-theoretical indices.
  • * Multivariate outlier testing to identify and address discordant observations.

Related Experiment Videos

  • * Comparison of PCA-derived scores with stepwise multiple regression models for predicting biological activity.

    • Main Results:

    • * PCA successfully identified correlations with biological activity, particularly in the second principal component after orthogonal projection.
    • * PCA scores demonstrated superior predictive accuracy for biological activity compared to multiple and stepwise regression models.
    • * Identification and removal of a multivariate outlier significantly enhanced the predictive performance.

    Conclusions:

    • * PCA is a powerful tool for uncovering hidden correlations in complex chemical datasets, outperforming standard regression.
    • * Multivariate outlier detection is crucial for improving the robustness and accuracy of QSAR models.
    • * The findings suggest that PCA-based approaches can enhance drug discovery efforts by providing more accurate predictions of biological activity.