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

Multiobjective optimization in quantitative structure-activity relationships: deriving accurate and interpretable

Orazio Nicolotti1, Valerie J Gillet, Peter J Fleming

  • 1Krebs Institute for Biomolecular Research and Department of Information Studies, University of Sheffield, Western Bank, United Kingdom.

Journal of Medicinal Chemistry
|November 1, 2002
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

Explainable Bidirectional Long Short-Term Memory Networks Learn Chemistry from SMILES for Predicting Toxicity of Androgen and Estrogen Receptor Chemicals.

ACS omega·2026
Same author

Developing Predictive Models by Sharing Predictions - An Investigation of a Federated Learning Approach for ADMET Predictions.

Journal of medicinal chemistry·2026
Same author

Targeting GPR18: Structural modelling, ligand discovery, and therapeutic potential in neuroinflammatory disorders.

Biochemical pharmacology·2026
Same author

Automated Molecular Design in BRADSHAW, Applied to the Optimization of ERAP1 Inhibitors.

Journal of medicinal chemistry·2026
Same author

Development of Styryl-Modified 3,4-Dihydropyrimidin-2(1H)-ones as Potential Antitumor Agents.

ChemMedChem·2026
Same author

Explicit Applicability Domain Calculations Can Help Determine When Uncertainty Estimates Are Less Reliable.

ACS omega·2026
Same journal

Rational Development of Activatable Prodrugs of the GSTP1 Inhibitor NBDHEX: Turn-On NIR Fluorogenic Drug Delivery with Selective Anticancer Activity.

Journal of medicinal chemistry·2026
Same journal

Recent Highlights in the Discovery and Design of Antibody-Drug Conjugates.

Journal of medicinal chemistry·2026
Same journal

One Face, Three Solutions: Structural Convergence in PD-L1 Inhibition across Antibodies, Macrocycles, and Small Molecules.

Journal of medicinal chemistry·2026
Same journal

A Potent and Selective Quinolone-Based PTPN22 Inhibitor with Improved Immunotherapeutic Activity.

Journal of medicinal chemistry·2026
Same journal

Structure-Based Discovery of (<i>E</i>)- and (<i>Z</i>)-4-Oxobut-2-enoate Derivatives as Novel Covalent Pin1 Inhibitors with Antitumor Activity.

Journal of medicinal chemistry·2026
Same journal

Hybrids of Benzenesulfonamide Oxadiazole Derivatives with Dual CA II and COX-2 Inhibitory Activity Demonstrating Antiglaucoma and Anti-inflammatory Action: Synthesis, In Silico Insights, and In Vitro and In Vivo Bioevaluation.

Journal of medicinal chemistry·2026
See all related articles

Two new genetic programming methods, genetic QSAR (GPQSAR) and multiobjective QSAR (MoQSAR), were developed to create accurate and interpretable quantitative structure-activity relationship models. MoQSAR offers a flexible approach, balancing statistical robustness with chemical interpretability for medicinal chemists.

Area of Science:

  • * Computational Chemistry
  • * Cheminformatics
  • * Machine Learning

Background:

  • * Deriving accurate, reliable, and interpretable quantitative structure-activity relationship (QSAR) models is challenging.
  • * Existing methods often struggle to balance model complexity and predictive power.
  • * Understanding biochemical responses requires chemically intuitive yet statistically robust models.

Purpose of the Study:

  • * To develop novel genetic programming (GP) based methods for QSAR model generation.
  • * To create models that balance accuracy, complexity, and interpretability.
  • * To introduce a 'chemical desirability' objective for enhanced biological understanding.

Main Methods:

  • * Genetic QSAR (GPQSAR): Employs a penalty function to control complexity, deriving a single linear model.

Related Experiment Videos

  • * Multiobjective Genetic QSAR (MoQSAR): Treats QSAR as a multiobjective optimization problem using multiobjective GP.
  • * MoQSAR incorporates objectives like model fitting, term count, nonlinearity, and chemical interpretability.

    • Main Results:

    • * Both GPQSAR and MoQSAR were tested on diverse datasets, including solubility data.
    • * MoQSAR successfully generated a family of QSAR models representing different objective trade-offs.
    • * The study demonstrated MoQSAR's ability to find models comparable to standard statistical approaches.

    Conclusions:

    • * MoQSAR provides a flexible framework for developing QSAR models with tunable statistical robustness and chemical interpretability.
    • * This approach aids medicinal chemists in selecting models that balance predictive performance with biological insight.
    • * The developed methods offer improved strategies for quantitative structure-activity relationship modeling.