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

Selecting genes by test statistics.

Dechang Chen1, Zhenqiu Liu, Xiaobin Ma

  • 1Division of Epidemiology and Biostatistics, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, USA. dchen@usuhs.mil

Journal of Biomedicine & Biotechnology
|July 28, 2005
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

Corrigendum to "The current situation of intensive care units in Chinese mainland: A nationwide survey" [Journal of Intensive Medicine volume 2 (2026) 132-140].

Journal of intensive medicine·2026
Same author

Ferritinophagy-Related Genes in Breast Cancer: Insights from Multi-Omics Analysis.

Current medicinal chemistry·2026
Same author

Video-Based Training on Physicians' Proficiency in Invasive Fungal Disease: A Multicentre, Cluster Randomised Controlled Trial.

Mycoses·2026
Same author

Photon-counting CT angiography for anatomical assessment of patent ductus arteriosus in a child.

Cardiology in the young·2026
Same author

Adaptive spatiotemporal graph learning for multi-horizon probabilistic wind power forecasting.

Scientific reports·2026
Same author

FKBP9 Enhances Colon Cancer Cell Proliferation by Inhibiting GPX4-Mediated Ferroptosis.

Cancer medicine·2026
Same journal

The effect of glycyrrhetinic acid on pharmacokinetics of cortisone and its metabolite cortisol in rats.

Journal of biomedicine & biotechnology·2012
Same journal

Insights and hopes in umbilical cord blood stem cell transplantations.

Journal of biomedicine & biotechnology·2012
Same journal

Three-dimensional visualization with large data sets: a simulation of spreading cortical depression in human brain.

Journal of biomedicine & biotechnology·2012
Same journal

Bioconversion of sugarcane biomass into ethanol: an overview about composition, pretreatment methods, detoxification of hydrolysates, enzymatic saccharification, and ethanol fermentation.

Journal of biomedicine & biotechnology·2012
Same journal

Trends in tissue engineering for blood vessels.

Journal of biomedicine & biotechnology·2012
Same journal

Salinomycin as a drug for targeting human cancer stem cells.

Journal of biomedicine & biotechnology·2012
See all related articles

This study evaluates alternative gene selection methods for multiclass microarray analysis, addressing limitations of the traditional F test by exploring statistics that handle unequal variances, improving classification accuracy.

Area of Science:

  • Bioinformatics
  • Statistical genomics
  • Gene expression analysis

Background:

  • Gene selection is crucial for analyzing multiclass microarray data.
  • The traditional ANOVA F test statistic, commonly used for gene selection, assumes equal variances, which is often violated in gene expression data.
  • This limitation can impact the accuracy of gene identification for classification and discovery.

Purpose of the Study:

  • To explore and evaluate alternative gene selection test statistics that can accommodate heterogeneity of variances in multiclass microarray data.
  • To compare the performance of these alternative statistics against the traditional F test statistic.
  • To assess the impact of different test statistics on various classification methods.

Main Methods:

  • Investigated five alternative test statistics, including the Brown-Forsythe and Welch test statistics.

Related Experiment Videos

  • Evaluated the performance of these statistics using publicly available microarray datasets.
  • Compared the selected genes and classification performance obtained using the alternative statistics versus the traditional F statistic.

    • Main Results:

    • Alternative test statistics demonstrated effectiveness in handling variance heterogeneity.
    • Performance comparisons indicated potential advantages of certain alternative statistics over the F test in specific scenarios.
    • The choice of test statistic influenced the identification of informative genes and subsequent classification accuracy.

    Conclusions:

    • Alternative test statistics offer a more robust approach to gene selection in multiclass microarray analysis when variances are unequal.
    • These methods can lead to improved identification of relevant genes and enhanced classification performance.
    • Further investigation into these alternative statistics is warranted for optimizing gene expression data analysis.