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

Distance based algorithms for small biomolecule classification and structural similarity search.

Emre Karakoc1, Artem Cherkasov, S Cenk Sahinalp

  • 1School of Computing Science, Simon Fraser University, Burnaby, BC, Canada. cenk@cs.sfu.ca

Bioinformatics (Oxford, England)
|July 29, 2006
PubMed
Summary

This study introduces an optimized weighted Minkowski distance (wL(p)) for k-nearest-neighbor (k-nn) search in drug discovery. This method enhances compound classification accuracy and speeds up similarity searches in large molecular databases.

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

Deep docking, part 2: an amplified DDU platform for ultra-large virtual screening.

Chemical science·2026
Same author

LOCALE: Local-Alignment Embeddings for Noise-Robust DNA Search at SRA Scale.

bioRxiv : the preprint server for biology·2026
Same author

Assessing diaphragm dysfunction in ICU patients with COVID-19 pneumonia: A prospective observational cohort study.

Science progress·2026
Same author

Identifying Robust Subclonal Structures through Tumor Progression Tree Alignment.

bioRxiv : the preprint server for biology·2026
Same author

LCPAN: efficient variation graph construction using locally consistent parsing.

Genome biology·2026
Same author

The Use of DeepQSAR Models for the Discovery of Peptides With Enhanced Antimicrobial and Antibiofilm Potential.

Molecular informatics·2026

Area of Science:

  • Computational chemistry
  • Chemoinformatics
  • Drug discovery

Background:

  • Structural similarity search is crucial for molecular classification and in-silico drug discovery.
  • Effective similarity search requires discriminating compound bioactivity and efficient database searching.

Purpose of the Study:

  • To computationally design an optimal weighted Minkowski distance (wL(p)) for maximizing discrimination between active and inactive compounds.
  • To develop efficient k-nearest-neighbor (k-nn) search data structures for wL(p) distances.
  • To improve the accuracy and speed of similarity searches in large molecular databases.

Main Methods:

  • Focus on the k-nearest-neighbor (k-nn) search method for small molecule classification.
  • Computationally design optimal weighted Minkowski distance (wL(p)) metrics.

Related Experiment Videos

  • Construct pruning-based k-nn search data structures for efficient similarity searches.

    • Main Results:

    • The developed classifier achieves higher accuracy than Linear Discriminant Analysis (LDA) and Multivariable Regression (MLR) methods.
    • Performance is comparable to Artificial Neural Network (ANN) methods in accuracy.
    • The classifier is significantly faster than ANN, especially for large datasets.
    • It quantifies bioactivity levels, offering more information than binary classification.

    Conclusions:

    • The optimized wL(p) distance and k-nn search strategy provide a powerful tool for in-silico drug discovery.
    • This approach offers a more accurate, faster, and informative method for molecular classification and bioactivity prediction.