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

Prediction of aqueous solubility and partition coefficient optimized by a genetic algorithm based descriptor

Jörg K Wegner1, Andreas Zell

  • 1Zentrum für Bioinformatik Tübingen, Universität Tübingen, Sand 1, D-72076 Tübingen, Germany. wegnerj@informatik.uni-tuebingen.de

Journal of Chemical Information and Computer Sciences
|May 28, 2003
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

Improving route development using convergent retrosynthesis planning.

Journal of cheminformatics·2025
Same author

Correction: Global reactivity models are impactful in industrial synthesis applications.

Journal of cheminformatics·2023
Same author

Global reactivity models are impactful in industrial synthesis applications.

Journal of cheminformatics·2023
Same author

Improving Few- and Zero-Shot Reaction Template Prediction Using Modern Hopfield Networks.

Journal of chemical information and modeling·2022
Same author

SBMLSimulator: A Java Tool for Model Simulation and Parameter Estimation in Systems Biology.

Computation (Basel, Switzerland)·2020
Same author

A genetics-led approach defines the drug target landscape of 30 immune-related traits.

Nature genetics·2019
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
Same journal

Future Papers.

Journal of chemical information and computer sciences·2016
See all related articles

This study introduces a genetic algorithm for descriptor selection (GA-SEC), using Shannon entropy to identify key molecular descriptors for accurate property prediction. The method efficiently builds predictive models for solubility and partition coefficients.

Area of Science:

  • Computational chemistry
  • Cheminformatics
  • Machine learning

Background:

  • Accurate prediction of physicochemical properties like aqueous solubility (logS) and octanol/water partition coefficient (logP) is crucial in drug discovery and chemical research.
  • Traditional descriptor selection methods can be computationally intensive and may not effectively handle large datasets.
  • Identifying a minimal set of relevant, interpretable descriptors is essential for building robust predictive models.

Purpose of the Study:

  • To develop and validate a fast and flexible descriptor selection method, termed GA-SEC (Genetic Algorithm for Selection of Essential Components).
  • To utilize Shannon entropy (SE) and differential Shannon entropy (DSE) for efficient descriptor relevance assessment, enabling the processing of large datasets.
  • To automatically build accurate predictive models using a small subset of the most important, transparent molecular descriptors.

Related Experiment Videos

Main Methods:

  • Implementation of a genetic algorithm variant (GA-SEC) for descriptor selection.
  • Application of Shannon entropy (SE) and differential Shannon entropy (DSE) to measure descriptor relevance and minimize memory footprint.
  • Development of artificial neural network (ANN) models for predicting aqueous solubility (logS) and octanol/water partition coefficient (logP).
  • Validation of the selected descriptors and predictive models on independent training and test datasets.

Main Results:

  • The GA-SEC method successfully identified a small set of pure, non-redundant descriptors.
  • ANN models built using GA-SEC selected descriptors achieved high predictive accuracy for logS (correlation coefficient = 0.93, empirical standard deviation = 0.54) and logP (correlation coefficient = 0.92, empirical standard deviation = 0.44).
  • The use of SE and DSE allowed for efficient processing of large chemical datasets.

Conclusions:

  • GA-SEC provides an efficient and effective approach for selecting relevant molecular descriptors for property prediction.
  • The identified descriptors are transparent and contribute to interpretable predictive models.
  • The method demonstrates significant potential for accelerating cheminformatics and drug discovery workflows by enabling rapid and accurate property prediction.