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

Sample size planning for developing classifiers using high-dimensional DNA microarray data.

Kevin K Dobbin1, Richard M Simon

  • 1Biometric Research Branch, National Cancer Institute, 6130 Executive Boulevard, Rockville, MD 20852, USA. dobbinke@mail.nih.gov

Biostatistics (Oxford, England)
|April 15, 2006
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

Exploring bias due to below-limit-of-detection values in influenza vaccine antibody modeling: A case study and instructional guide for the CIVIC study.

Vaccine·2025
Same author

Human basal-like breast cancer is represented by one of the two mammary tumor subtypes in dogs.

Breast cancer research : BCR·2023
Same author

Human basal-like breast cancer is represented by one of the two mammary tumor subtypes in dogs.

bioRxiv : the preprint server for biology·2023
Same author

Development and internal validation of risk scores to diagnose infectious mononucleosis among college students.

Family practice·2022
Same author

Access and Utilization of Asthma Medications Among Patients Who Receive Care in Federally Qualified Health Centers.

Journal of primary care & community health·2022
Same author

Sample size methods for evaluation of predictive biomarkers.

Statistics in medicine·2022

Developing accurate gene expression predictors requires careful sample size calculation. This study presents a new method accounting for gene selection variability, suggesting many prediction tasks need smaller training datasets than anticipated.

Area of Science:

  • Bioinformatics
  • Statistical Genetics
  • Machine Learning in Biology

Background:

  • Gene expression studies often aim to classify samples into diagnostic or prognostic groups.
  • When biological classes are similar, few genes differentiate them, necessitating gene selection in predictor development.
  • Both gene selection and predictor construction introduce variability impacting classifier accuracy.

Purpose of the Study:

  • To introduce a methodology for sample size determination in high-dimensional gene expression prediction.
  • To account for variability introduced by both gene subset selection and classifier construction.
  • To provide a practical approach for sample size calculation without extensive preliminary data.

Main Methods:

  • Developed a parametric probability model for sample size estimation.

Related Experiment Videos

  • Incorporated variability from both gene selection and predictor development steps.
  • Enabled practical sample size computations for high-dimensional data.

    • Main Results:

    • The proposed methodology captures key sources of variability in predictor development.
    • Sample size requirements can be practically computed using the developed model.
    • Many gene expression prediction problems may not necessitate large training datasets.

    Conclusions:

    • Accurate sample size estimation is crucial for reliable gene expression-based predictors.
    • The new methodology offers a robust framework for sample size determination in high-dimensional settings.
    • The findings suggest potential for more efficient study designs in gene expression-based classification.