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

Self-organizing neural networks for modeling 3D QSAR--a comparative study.

J Polański1, B Walczak, R Gieleciak

  • 1Institute of Chemistry, University of Silesia, PL-40-006 Katowice, Poland. Polanskiùs.edu.pl

Acta Poloniae Pharmaceutica
|April 11, 2001
PubMed
Summary

Self-organizing neural networks (SOM) effectively model 3D Quantitative Structure-Activity Relationships (QSAR) using atomic data. Testing on steroid-corticosteroid binding globulin complexes revealed architectural sensitivities crucial for drug design performance.

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

Can life satisfaction be considered a predictor of quality of life in patients with lung cancer?

European review for medical and pharmacological sciences·2020
Same author

What can go wrong at the data normalization step for identification of biomarkers?

Journal of chromatography. A·2014
Same author

Synergy against fungal pathogens: working together is better than working alone.

Current medicinal chemistry·2013
Same author

High-dimensional nested analysis of variance to assess the effect of production season, quality grade and steam pasteurization on the phenolic composition of fermented rooibos herbal tea.

Talanta·2013
Same author

Elemental and spectroscopic characterization of fractions of an acidic extract of oil sands process water.

Chemosphere·2013
Same author

FTIR spectroscopic characterization of differently cultivated food related yeasts.

The Analyst·2013

Area of Science:

  • Computational chemistry
  • Cheminformatics
  • Artificial intelligence in drug discovery

Background:

  • 3D Quantitative Structure-Activity Relationship (QSAR) modeling is vital for predicting drug efficacy.
  • Self-organizing neural networks (SOM) offer a robust framework for complex data analysis.
  • Evaluating different SOM architectures is essential for optimizing predictive models.

Purpose of the Study:

  • To investigate the performance of various self-organizing neural network (SOM) architectures for 3D QSAR modeling.
  • To assess the impact of molecular alignment strategies on SOM-based drug design predictions.
  • To compare the efficacy of different SOM methods using a benchmark dataset of steroids complexing corticosteroid binding globulin (CBG).

Main Methods:

  • Utilized atomic coordinates and partial atomic charges as input features for SOM models.

Related Experiment Videos

  • Employed a benchmark dataset of steroid-CBG complexes for comparative analysis.
  • Tested the sensitivity of different SOM architectures to variations in molecular alignment, including alignment based on molecular inertial axes.

    • Main Results:

    • Demonstrated that specific SOM architectures exhibit varying sensitivities to molecular alignment.
    • Showcased the influence of alignment strategies on the predictive performance of 3D QSAR models.
    • Identified optimal SOM configurations for modeling the interactions between steroids and CBG.

    Conclusions:

    • Different SOM architectures possess distinct strengths and weaknesses for 3D QSAR modeling.
    • Accurate molecular alignment is a critical factor for successful SOM-based drug design.
    • The findings provide valuable insights for selecting and optimizing SOM methods in computational drug discovery.