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

Partitioning in binary-transformed chemical descriptor spaces.

Jeffrey W Godden1, Jürgen Bajorath

  • 1Computer Aided Drug Discovery, Albany Molecular Research Inc., Bothell Research Center, Washington, USA.

Methods in Molecular Biology (Clifton, N.J.)
|May 14, 2004
PubMed
Summary

Median partitioning (MP) transforms molecular property data into binary classifications. This statistically based method efficiently partitions large molecular datasets without dimension reduction, proving effective for diversity selection and virtual screening.

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

Global Assessment of Substituents on the Basis of Analogue Series.

Journal of medicinal chemistry·2020
Same author

From Qualitative to Quantitative Analysis of Activity and Property Landscapes.

Journal of chemical information and modeling·2020
Same author

Systematic Data Analysis and Diagnostic Machine Learning Reveal Differences between Compounds with Single- and Multitarget Activity.

Molecular pharmaceutics·2020
Same author

Data set of activity cliffs with single-atom modification and associated X-ray structure information for medicinal and computational chemistry applications.

Data in brief·2020
Same author

DeepCOMO: from structure-activity relationship diagnostics to generative molecular design using the compound optimization monitor methodology.

Journal of computer-aided molecular design·2020
Same author

Inhibitor bias in luciferase-based luminescence assays.

Future science OA·2020

Area of Science:

  • Computational chemistry
  • cheminformatics
  • data science

Background:

  • Traditional partitioning methods for molecular datasets often rely on dimension reduction techniques.
  • These methods can be computationally intensive when analyzing large compound databases.

Purpose of the Study:

  • To introduce and evaluate a statistically based partitioning method called median partitioning (MP).
  • To demonstrate that effective molecular dataset partitioning can be achieved without dimension reduction.

Main Methods:

  • Median partitioning (MP) transforms molecular property descriptor value distributions into a binary classification scheme.
  • MP operates directly in the original, simplified chemical space, unlike dimension reduction approaches.
  • Modified MP algorithms were applied to diversity selection, compound classification, and virtual screening tasks.

Related Experiment Videos

Main Results:

  • MP provides an effective alternative to dimension reduction techniques for partitioning molecular datasets.
  • Modified MP algorithms showed successful application in diversity selection, compound classification, and virtual screening.
  • MP demonstrates significant computational efficiency, crucial for analyzing large-scale compound databases.

Conclusions:

  • Dimension reduction is not essential for effective partitioning of molecular datasets.
  • Statistically based partitioning methods like MP offer computational efficiency for large-scale data analysis.
  • MP is a valuable tool for cheminformatics tasks such as virtual screening and compound classification.