Jove
Visualize
Contact Us

Related Experiment Video

Updated: May 24, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Finding biomarkers is getting easier.

Brian Patrick Bradley1

  • 1Department of Biological Sciences, University of Maryland Baltimore County, Baltimore, MD 21250, USA. bbradley@umbc.edu

Ecotoxicology (London, England)
|March 14, 2012
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 journal

In vivo effects of lead on sea urchin Paracentrotus lividus: toxicity, bioconcentration and oxidative stress response.

Ecotoxicology (London, England)·2026
Same journal

Morphological and toxicological effects of a deltamethrin-based insecticide on aquatic insect larvae: Grumichella boraceia (Trichoptera) as a bioindicator of pyrethroid contamination.

Ecotoxicology (London, England)·2026
Same journal

Oxidative stress response biomarkers in gills and liver of bullfrog tadpoles semi-chronically exposed to combined microplastics and titanium dioxide nanoparticles.

Ecotoxicology (London, England)·2026
Same journal

Ecotoxicity evaluation of thymol in tunneller dung beetles (Coleoptera: Geotrupidae) as a natural alternative to ecotoxic anthelmintics.

Ecotoxicology (London, England)·2026
Same journal

Neonicotinoid insecticide thiamethoxam compromises early larval development of the Amazonian tetra fish Astyanax bimaculatus (Linnaeus, 1758).

Ecotoxicology (London, England)·2026
Same journal

Response of soil microbiomes to nano-zero-valent iron and biochar in Cr(VI)-contaminated soil remediation.

Ecotoxicology (London, England)·2026
See all related articles
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

Machine learning (ML) identifies potent protein biomarker combinations for accurate diagnostics. This approach overcomes limitations of single biomarkers and suites, enabling precise identification of analytes and toxicity levels.

Area of Science:

  • Biochemistry
  • Bioinformatics
  • Computational Biology

Background:

  • Single biomarkers and biomarker suites often yield inaccurate or conflicting results.
  • Identifying potent combinations of variables is crucial for accurate analyte and toxicity level identification.

Purpose of the Study:

  • To review how machine learning (ML), particularly artificial neural networks, can identify potent biomarkers from expression data.
  • To demonstrate the effectiveness of ML in reducing thousands of candidate variables to a small set for treatment classification.

Main Methods:

  • Utilizing machine learning, specifically artificial neural networks, to analyze expression data, primarily proteins.
  • Employing iterative computer processes to systematically identify sets of proteins that classify treatment groups.

More Related Videos

Dried Blood Spot Collection of Health Biomarkers to Maximize Participation in Population Studies
07:20

Dried Blood Spot Collection of Health Biomarkers to Maximize Participation in Population Studies

Published on: January 28, 2014

Related Experiment Videos

Last Updated: May 24, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Dried Blood Spot Collection of Health Biomarkers to Maximize Participation in Population Studies
07:20

Dried Blood Spot Collection of Health Biomarkers to Maximize Participation in Population Studies

Published on: January 28, 2014

Main Results:

  • ML successfully identifies key protein biomarkers with near 100% accuracy for specific treatment classes.
  • The identified biomarkers show potential for portable field tests for various adverse conditions.

Conclusions:

  • Machine learning offers a powerful method for discovering effective diagnostic biomarkers, especially proteins.
  • While ML facilitates biomarker discovery, confirmation through multivariable statistics and field studies remains essential.