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

Extending the trend vector: the trend matrix and sample-based partial least squares

R P Sheridan1, R B Nachbar, B L Bush

  • 1Molecular Systems Department, Merck Research Laboratories, Rahway, NJ 07065.

Journal of Computer-Aided Molecular Design
|June 1, 1994
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

Protocols for bridging the peptide to nonpeptide gap in topological similarity searches.

Journal of chemical information and computer sciences·2001
Same author

Comparison of knowledge-based and distance geometry approaches for generation of molecular conformations.

Journal of chemical information and computer sciences·2001
Same author

Mining the chemical quarry with joint chemical probes: an application of latent semantic structure indexing (LaSSI) and TOPOSIM (Dice) to chemical database mining.

Journal of medicinal chemistry·2001
Same author

Latent semantic structure indexing (LaSSI) for defining chemical similarity.

Journal of medicinal chemistry·2001
Same author

Chemical similarity searches using latent semantic structural indexing (LaSSI) and comparison to TOPOSIM.

Journal of medicinal chemistry·2001
Same author

Designing targeted libraries with genetic algorithms.

Journal of molecular graphics & modelling·2001
Same journal

N2-substituted triazoles as trypanothione reductase inhibitors: in silico evaluation, synthesis, and anti-leishmanial activity.

Journal of computer-aided molecular design·2026
Same journal

Theoretical study for investigating three experimental dose-response curves of rosy odorants on human olfactory receptor OR2A25 via molecular docking calculations and statistical physics modeling.

Journal of computer-aided molecular design·2026
Same journal

A linear models approach to optimize carbazole-based dyes for solar cell applications.

Journal of computer-aided molecular design·2026
Same journal

DOCKweb: a web-based GUI platform for molecular modeling with DOCK6.

Journal of computer-aided molecular design·2026
Same journal

Two faces of drug repurposing data in machine learning-driven structure-based virtual screening supported by interaction fingerprints.

Journal of computer-aided molecular design·2026
Same journal

Repurposing bleomycin against Acinetobacter baumannii HisG: computational, biophysical, and antibacterial evidence.

Journal of computer-aided molecular design·2026
See all related articles

Trend vector analysis in drug discovery improves compound ranking. New trend matrix and SAMPLS methods enhance predictions, with SAMPLS offering superior efficiency and visualization capabilities for predicting biological activity.

Area of Science:

  • Computational Chemistry
  • Cheminformatics
  • Drug Discovery

Background:

  • Trend vector analysis, using topological descriptors, aids in ranking chemical compounds by predicted biological activity.
  • Existing methods can be enhanced to improve the accuracy of these predictions in drug discovery pipelines.

Purpose of the Study:

  • To generalize trend vector analysis for improved prediction of biological activity ranking.
  • To introduce and evaluate the trend matrix and SAMPLS methods for enhanced predictive modeling.
  • To develop statistical checks to prevent overfitting in predictive models.

Main Methods:

  • Trend matrix method: Analyzes correlations between residuals and simultaneous descriptor occurrences.
  • SAMPLS method: Utilizes partial least squares (PLS) for efficient linear model derivation with numerous descriptors.

Related Experiment Videos

  • Statistical checks and randomization methods were employed to prevent overfitting and determine optimal PLS components.

    • Main Results:

    • Both trend matrix and SAMPLS methods improved prediction accuracy over the simple trend vector.
    • SAMPLS demonstrated superior performance with reduced storage and CPU time requirements.
    • SAMPLS provided useful axes for visualizing compound properties and enhanced prediction of activities in test sets.

    Conclusions:

    • Generalizing trend vector analysis with methods like trend matrix and SAMPLS significantly enhances predictive accuracy in drug discovery.
    • SAMPLS is a highly efficient and effective method for large-scale compound activity prediction and property visualization.
    • Statistical validation is crucial to mitigate overfitting risks when employing advanced predictive modeling techniques.