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

Database composition can affect the structure-activity relationship prediction.

John F Young1, Chen-An Tsai, James J Chen

  • 1Division of Biometry and Risk Assessment, National Center for Toxicological Research, Food and Drug Administration, Jefferson, Arkansas 72079-9502, USA. jyoung@nctr.fda.gov

Journal of Toxicology and Environmental Health. Part A
|July 21, 2006
PubMed
Summary

The ratio of active to inactive chemicals impacts structure-activity relationship (SAR) analysis sensitivity and specificity. Increasing active chemicals boosts sensitivity but lowers specificity, with concordance remaining stable.

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

Efficient Selection of Gaussian Kernel SVM Parameters for Imbalanced Data.

Genes·2023
Same author

A molecular substitutional disorder case study suitable for instruction: L<sub>2</sub>Cr<sup>II</sup>(THF)/L<sub>2</sub>[(trimethylsilyl)methyl]Cr<sup>III</sup> (L is 2,5-bis{[(2,6-diisopropylphenyl)imino]methyl}pyrrol-1-ide).

Acta crystallographica. Section C, Structural chemistry·2022
Same author

Echocardiography and Management for Cardiac Trauma.

Journal of cardiothoracic and vascular anesthesia·2022
Same author

Effects of intravenous and oral di(2-ethylhexyl) phthalate (DEHP) and 20% Intralipid vehicle on neonatal rat testis, lung, liver, and kidney.

Food and chemical toxicology : an international journal published for the British Industrial Biological Research Association·2020
Same author

Prognostic and predictive signatures for treatment decisions.

Biomarkers in medicine·2018
Same author

Difficult Retrieval of a Broken Catheter From a Total Implantable Venous Access Device.

The Journal of invasive cardiology·2018

Area of Science:

  • Toxicology
  • Computational Chemistry
  • Bioinformatics

Background:

  • Structure-activity relationship (SAR) analyses are crucial for predicting chemical properties.
  • The performance of SAR models, specifically sensitivity and specificity, can be influenced by the proportion of active and inactive chemicals in the training dataset.
  • Understanding this influence is key to developing robust predictive models.

Purpose of the Study:

  • To investigate how the ratio of active (A) to inactive (I) chemicals affects the sensitivity and specificity of SAR analyses.
  • To determine the impact of dataset size on these predictive metrics.

Main Methods:

  • The National Center for Toxicological Research (NCTR) liver cancer database (NCTRlcdb) was utilized.
  • The database was subdivided into datasets with varying A/I ratios, ranging from 0.2 to 5.5.

Related Experiment Videos

  • Sensitivity, specificity, and concordance were calculated for each subset, with dataset sizes ranging from 187 to 999 chemicals.

    • Main Results:

    • A significant variation in sensitivity/specificity ratios (0.1 to 6.5) was observed across different A/I ratios.
    • As the proportion of active chemicals increased (higher A/I ratio), sensitivity consistently rose.
    • Conversely, specificity decreased as the A/I ratio increased, while overall concordance remained relatively stable.
    • Dataset size (187-999 chemicals) did not demonstrably affect sensitivity, specificity, or concordance.

    Conclusions:

    • The A/I ratio is a critical factor influencing the predictive performance of SAR models.
    • Optimizing the A/I ratio can enhance model sensitivity, though it may necessitate a trade-off with specificity.
    • Dataset size appears to be a less critical factor than the A/I ratio in determining SAR model performance for this dataset.