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

Variable selection and interpretation in structure-affinity correlation modeling of estrogen receptor binders.

Federico Marini1, Alessandra Roncaglioni, Marjana Novic

  • 1National Institute of Chemistry, Ljubljana, Slovenia, University of Rome La Sapienza, Rome, Italy.

Journal of Chemical Information and Modeling
|November 29, 2005
PubMed
Summary

This study developed computational models to predict estrogen receptor binding affinity. A nonlinear neural network model achieved the highest predictive accuracy, offering insights into estrogenic compound interactions.

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

Non-destructive forensic identification and quantification of cocaine through sealed packaging using FT-NIR spectroscopy and hyperspectral imaging.

Analytica chimica acta·2026
Same author

Type I interferon-activated NK cells control polycythemia vera in vivo.

Blood·2026
Same author

TGF-β1 and TGF-β2 family members differentially modulate tumor initiation and invasiveness of primary liver cancer in a MMP14-dependent manner.

Carcinogenesis·2026
Same author

Beyond Molecular Structures: Investigating Demographic Factors in Drug-Induced Cardiotoxicity Prediction Models.

Journal of chemical information and modeling·2026
Same author

Simulation of fish chronic toxicity using the Las Vegas algorithm and the vector of ideality of correlation.

Environmental toxicology and chemistry·2026
Same author

The MYB-related transcription factor MYPOP acts as a selective regulator of cancer cell growth.

Communications biology·2026

Area of Science:

  • Computational chemistry
  • Molecular modeling
  • Pharmacology

Background:

  • Understanding estrogen receptor (ER) binding is crucial for drug development.
  • Predicting chemical structure-activity relationships (SAR) aids in identifying potent ER modulators.
  • Existing methods require efficient computational approaches for SAR analysis.

Purpose of the Study:

  • To develop and compare computational models for predicting estrogen receptor binding affinity.
  • To identify key structural descriptors influencing ER binding.
  • To gain insights into the mechanisms of estrogenic compound interactions with the ER.

Main Methods:

  • Utilized a dataset of 132 compounds with known structures and ER binding affinities.
  • Applied multivariate modeling techniques: partial least-squares regression (PLSR), counterpropagation neural network (CPNN), and error-back-propagation neural network (EBPNN).

Related Experiment Videos

  • Employed variable selection methods, including "variable importance in projection" for PLSR and genetic algorithms (GA) for neural networks, followed by leave-one-out cross-validation (LOOCV).

    • Main Results:

    • Compared the predictive abilities of PLSR, CPNN, and EBPNN models.
    • Genetic algorithms effectively selected optimal structural descriptors for neural network models.
    • The EBPNN model demonstrated superior predictive performance with R²=92.2% and Q²=70.8% after LOOCV.

    Conclusions:

    • Computational modeling, particularly nonlinear neural networks, can accurately predict estrogen receptor binding affinity.
    • Variable selection using genetic algorithms enhances model interpretability and predictive power.
    • The findings provide valuable insights into the molecular mechanisms governing estrogenic compound interactions with the estrogen receptor.