Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Calculating confidence intervals for prediction error in microarray classification using resampling.

Wenyu Jiang¹, Sudhir Varma, Richard Simon

¹Concordia University, Quebec, Canada. wjiang@mathstat.concordia.ca

Statistical Applications in Genetics and Molecular Biology

|March 4, 2008

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 the impact of AI tools on filmmakers' creative satisfaction: the mediating role of creative efficiency and autonomy.

Frontiers in psychology·2026

Same author

Evaluation of periodontal ligament stem cell viability and proliferation on native collagen membranes: An <i>in vitro</i> study.

Journal of Indian Society of Periodontology·2026

Same author

2D Ultrasound Elasticity Imaging of Abdominal Aortic Aneurysms Using Deep Neural Networks.

IEEE transactions on computational imaging·2026

Same author

Wall Thinning Monitoring in Boiler U-Bends: A Review and Future Prospects with Fiber Optic Sensing.

Micromachines·2026

Same author

FBXL6 promotes bladder cancer progression by stabilizing ENO1 through K63-linked ubiquitination.

Cell death discovery·2026

Same author

Reversible Diode Effect in Truncated-pyramid-shaped Bi<sub>2</sub>O<sub>2</sub>Se Modulated by Tip-Force-Induced Flexoelectric Field.

ACS nano·2026

Same journal

Annealed variational mixtures for disease subtyping and biomarker discovery.

Statistical applications in genetics and molecular biology·2026

Same journal

Performance of the permutation test approach with base calling errors for detecting changes in variant allele frequencies in ctDNA for a single patient.

Statistical applications in genetics and molecular biology·2026

Same journal

BLOG: Bayesian longitudinal omics with group constraints.

Statistical applications in genetics and molecular biology·2026

Same journal

AI-driven risk prediction and categorization in cystic fibrosis leveraging AttentiveLSTM and Fox Wolf Optimizer.

Statistical applications in genetics and molecular biology·2026

Same journal

Perfect collinearity not created equal: measuring and visualizing the severity of multi-collinearity of modern omics data.

Statistical applications in genetics and molecular biology·2026

Same journal

Corrigendum to: Choice of baseline hazards in joint modeling of longitudinal and time-to-event cancer survival data.

Statistical applications in genetics and molecular biology·2025

See all related articles

Accurate prediction accuracy confidence intervals are crucial for microarray class prediction, especially with small sample sizes. A novel bias-reduced bootstrap case cross-validation method offers reliable confidence intervals for prediction accuracy in such scenarios.

Area of Science:

Bioinformatics
Statistical Learning
Computational Biology

Background:

Point estimates of prediction accuracy from cross-validation are common in microarray class prediction.
These estimates exhibit high variability, particularly with small sample sizes, necessitating confidence intervals.

Purpose of the Study:

To evaluate existing confidence interval methods for prediction accuracy in microarray studies.
To develop and validate a novel resampling scheme for improved confidence interval estimation.

Main Methods:

Extensive comparison of existing confidence interval methods based on empirical coverage and width.
Development of a bootstrap case cross-validation (BCCV) resampling scheme.
Definition and evaluation of BCCV-based confidence intervals, including bias-corrected and percentile methods.

Related Experiment Videos

Main Results:

Standard binomial assumption and split-sample methods show under-coverage or overly conservative intervals.
BCCV percentile intervals are conservative without bias-correction; Efron's BCa intervals are anti-conservative.
A simple bias reduction on the BCCV percentile interval yields mildly conservative and superior inference.

Conclusions:

Existing methods for confidence intervals of prediction accuracy in microarrays are inadequate, especially for small sample sizes.
The proposed bias-reduced BCCV percentile method provides reliable and more accurate confidence intervals.
This method outperforms others for small to moderate sample size microarray applications.