Split-plot microarray experiments: Issues of design, power and sample size

Pi Wen Tsai, Mei Ling Ting Lee

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

This article focuses on microarray experiments with two or more factors in which treatment combinations of the factors corresponding to the samples paired together onto arrays are not completely random. A main effect of one (or more) factor(s) is confounded with arrays (the experimental blocks). This is called a split-plot microarray experiment. We utilise an analysis of variance (ANOVA) model to assess differentially expressed genes for between-array and within-array comparisons that are generic under a split-plot microarray experiment. Instead of standard t- or F-test statistics that rely on mean square errors of the ANOVA model, we use a robust method, referred to as 'a pooled percentile estimator', to identify genes that are differentially expressed across different treatment conditions. We illustrate the design and analysis of split-plot microarray experiments based on a case application described by Jin et al. A brief discussion of power and sample size for split-plot microarray experiments is also presented.

Original languageEnglish
Pages (from-to)187-194
Number of pages8
JournalApplied Bioinformatics
Volume4
Issue number3
DOIs
Publication statusPublished - 2005 Jan 1

Fingerprint

Microarrays
Sample Size
Design of experiments
Analysis of Variance
analysis of variance
Genes
Analysis of variance (ANOVA)
genes
statistics
Experiments
sampling
Mean square error
testing
Statistics
methodology

ASJC Scopus subject areas

  • Information Systems
  • Agricultural and Biological Sciences(all)
  • Computer Science Applications

Cite this

Split-plot microarray experiments : Issues of design, power and sample size. / Tsai, Pi Wen; Lee, Mei Ling Ting.

In: Applied Bioinformatics, Vol. 4, No. 3, 01.01.2005, p. 187-194.

Research output: Contribution to journalArticle

@article{f51aa009080e4a90a6fcf90a1e3abec5,
title = "Split-plot microarray experiments: Issues of design, power and sample size",
abstract = "This article focuses on microarray experiments with two or more factors in which treatment combinations of the factors corresponding to the samples paired together onto arrays are not completely random. A main effect of one (or more) factor(s) is confounded with arrays (the experimental blocks). This is called a split-plot microarray experiment. We utilise an analysis of variance (ANOVA) model to assess differentially expressed genes for between-array and within-array comparisons that are generic under a split-plot microarray experiment. Instead of standard t- or F-test statistics that rely on mean square errors of the ANOVA model, we use a robust method, referred to as 'a pooled percentile estimator', to identify genes that are differentially expressed across different treatment conditions. We illustrate the design and analysis of split-plot microarray experiments based on a case application described by Jin et al. A brief discussion of power and sample size for split-plot microarray experiments is also presented.",
author = "Tsai, {Pi Wen} and Lee, {Mei Ling Ting}",
year = "2005",
month = "1",
day = "1",
doi = "10.2165/00822942-200504030-00003",
language = "English",
volume = "4",
pages = "187--194",
journal = "Applied Bioinformatics",
issn = "1175-5636",
publisher = "Adis Press",
number = "3",

}

TY - JOUR

T1 - Split-plot microarray experiments

T2 - Issues of design, power and sample size

AU - Tsai, Pi Wen

AU - Lee, Mei Ling Ting

PY - 2005/1/1

Y1 - 2005/1/1

N2 - This article focuses on microarray experiments with two or more factors in which treatment combinations of the factors corresponding to the samples paired together onto arrays are not completely random. A main effect of one (or more) factor(s) is confounded with arrays (the experimental blocks). This is called a split-plot microarray experiment. We utilise an analysis of variance (ANOVA) model to assess differentially expressed genes for between-array and within-array comparisons that are generic under a split-plot microarray experiment. Instead of standard t- or F-test statistics that rely on mean square errors of the ANOVA model, we use a robust method, referred to as 'a pooled percentile estimator', to identify genes that are differentially expressed across different treatment conditions. We illustrate the design and analysis of split-plot microarray experiments based on a case application described by Jin et al. A brief discussion of power and sample size for split-plot microarray experiments is also presented.

AB - This article focuses on microarray experiments with two or more factors in which treatment combinations of the factors corresponding to the samples paired together onto arrays are not completely random. A main effect of one (or more) factor(s) is confounded with arrays (the experimental blocks). This is called a split-plot microarray experiment. We utilise an analysis of variance (ANOVA) model to assess differentially expressed genes for between-array and within-array comparisons that are generic under a split-plot microarray experiment. Instead of standard t- or F-test statistics that rely on mean square errors of the ANOVA model, we use a robust method, referred to as 'a pooled percentile estimator', to identify genes that are differentially expressed across different treatment conditions. We illustrate the design and analysis of split-plot microarray experiments based on a case application described by Jin et al. A brief discussion of power and sample size for split-plot microarray experiments is also presented.

UR - http://www.scopus.com/inward/record.url?scp=27344457150&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27344457150&partnerID=8YFLogxK

U2 - 10.2165/00822942-200504030-00003

DO - 10.2165/00822942-200504030-00003

M3 - Article

C2 - 16231960

AN - SCOPUS:27344457150

VL - 4

SP - 187

EP - 194

JO - Applied Bioinformatics

JF - Applied Bioinformatics

SN - 1175-5636

IS - 3

ER -