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

Predictive toxicology: benchmarking molecular descriptors and statistical methods.

Jun Feng1, Laura Lurati, Haojun Ouyang

  • 1University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, USA.

Journal of Chemical Information and Computer Sciences
|September 23, 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

Onasemnogene abeparvovec gene therapy for symptomatic infantile-onset spinal muscular atrophy type 1 (STR1VE-EU): an open-label, single-arm, multicentre, phase 3 trial.

The Lancet. Neurology·2021
Same author

Onasemnogene abeparvovec gene therapy for symptomatic infantile-onset spinal muscular atrophy in patients with two copies of SMN2 (STR1VE): an open-label, single-arm, multicentre, phase 3 trial.

The Lancet. Neurology·2021
Same author

Evaluation of incomplete blinking as a measurement of dry eye disease.

The ocular surface·2019
Same author

CRISPR/Cas9-mediated efficient genome editing via protoplast-based transformation in yeast-like fungus Aureobasidium pullulans.

Gene·2019
Same author

Mechanically Reinforced Catechol-Containing Hydrogels with Improved Tissue Gluing Performance.

Biomimetics (Basel, Switzerland)·2019
Same author

<sup>68</sup>Ga-Pentixafor PET/CT for Imaging of Chemokine Receptor 4 Expression in Waldenström Macroglobulinemia/Lymphoplasmacytic Lymphoma: Comparison to <sup>18</sup>F-FDG PET/CT.

Journal of nuclear medicine : official publication, Society of Nuclear Medicine·2019

Predicting drug toxicity from chemical structure is crucial for efficient drug development. This study explored computational methods, finding success in smaller datasets but highlighting the need for new approaches for larger datasets to improve predictive toxicology.

Area of Science:

  • Computational Chemistry
  • Toxicology
  • Drug Discovery

Background:

  • Drug development requires identifying compounds with efficacy and minimal toxicity.
  • Assessing toxic effects is resource-intensive, necessitating predictive methods.
  • Predictive toxicology is complex due to multiple underlying toxic mechanisms.

Purpose of the Study:

  • To investigate the effectiveness of combining chemical descriptors and statistical methods for predictive toxicology.
  • To evaluate different computational approaches for predicting compound toxicity.
  • To identify limitations and areas for improvement in current predictive toxicology models.

Main Methods:

  • Collected and preprocessed four distinct toxicological datasets.
  • Calculated four diverse sets of chemical descriptors for each compound.

Related Experiment Videos

  • Applied three statistical modeling techniques: recursive partitioning, neural networks, and partial least squares.

    • Main Results:

    • Achieved accurate toxicity predictions on smaller datasets.
    • Observed reduced effectiveness of the methods on larger datasets.
    • Indicated that current chemical descriptors and statistical methods may be insufficient for large-scale predictive toxicology.

    Conclusions:

    • The study demonstrates the potential of computational methods in predictive toxicology.
    • Success varied with dataset size, suggesting a need for novel descriptors or advanced statistical techniques for larger datasets.
    • Further research is required to enhance understanding of descriptor-method interactions and develop improved predictive models.