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

A consensus neural network-based technique for discriminating soluble and poorly soluble compounds.

David T Manallack1, Benjamin G Tehan, Emanuela Gancia

  • 1Celltech R&D Ltd., Granta Park, Great Abington, Cambridge, CB1 6GS, United Kingdom. David.Manallack@denovopharma.com

Journal of Chemical Information and Computer Sciences
|March 26, 2003
PubMed
Summary

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

Intracrine FFA4 signaling controls lipolysis at lipid droplets.

Nature chemical biology·2025
Same author

Multicolored sequential resonance energy transfer for detection of simultaneous ligand binding at G protein-coupled receptors.

Nature communications·2025
Same author

Real-World Applications and Experiences of AI/ML Deployment for Drug Discovery.

Journal of medicinal chemistry·2025
Same author

Biased constitutive signaling of the G protein-coupled receptor GPR35 suppresses gut barrier permeability.

The Journal of biological chemistry·2024
Same author

The discovery of hidden guanylate cyclases (GCs) in the <i>Homo sapiens</i> proteome.

Computational and structural biotechnology journal·2023
Same author

Identification and Optimization of RNA-Splicing Modulators as Huntingtin Protein-Lowering Agents for the Treatment of Huntington's Disease.

Journal of medicinal chemistry·2023

This study uses Burden, CAS, and University of Texas (BCUT) descriptors and neural networks to predict compound solubility, achieving 95% accuracy in identifying poorly soluble compounds for drug discovery.

Area of Science:

  • Computational chemistry
  • Medicinal chemistry
  • Drug discovery

Background:

  • Burden, CAS, and University of Texas (BCUT) descriptors are eigenvalues of modified connectivity matrices used in drug design.
  • Traditionally, BCUT descriptors aid in selecting compounds by defining receptor-relevant subspaces.
  • Predicting aqueous solubility is crucial for drug formulation and discovery.

Purpose of the Study:

  • To develop a consensus neural network model using BCUT descriptors to differentiate compounds with poor aqueous solubility.
  • To establish a solubility threshold of 0.1 mg/mL for classifying compounds.
  • To improve the efficiency of candidate selection in drug discovery pipelines.

Main Methods:

  • Training consensus neural networks on BCUT descriptors.

Related Experiment Videos

  • Utilizing strict criteria for predicting insolubility.
  • Examining additional parameters for compounds with lower prediction confidence.

    • Main Results:

    • Achieved approximately 95% correct classification for compounds with poor aqueous solubility.
    • Successfully identified compounds with unsuitable biopharmaceutical and physicochemical properties.
    • Demonstrated the effectiveness of BCUT descriptors in solubility prediction.

    Conclusions:

    • The developed model serves as an effective filter for selecting screening candidates, guiding compound purchases, and prioritizing synthetic efforts.
    • This approach enhances the early stages of drug discovery by flagging compounds with potential solubility issues.
    • Integration of BCUT-based solubility prediction can streamline the development of new therapeutics.