Determination of minimum sample size and discriminatory expression patterns in microarray data

Daehee Hwang, William A. Schmitt, George Stephanopoulos, Gregory Stephanopoulos

Research output: Contribution to journalArticle

94 Citations (Scopus)

Abstract

Motivation: Transcriptional profiling using microarrays can reveal important information about cellular and tissue expression phenotypes, but these measurements are costly and time consuming. Additionally, tissue sample availability poses further constraints on the number of arrays that can be analyzed in connection with a particular disease or state of interest. It is therefore important to provide a method for the determination of the minimum number of microarrays required to separate, with statistical reliability, distinct disease states or other physiological differences. Results: Power analysis was applied to estimate the minimum sample size required for two-class and multi-class discrimination. The power analysis algorithm calculates the appropriate sample size for discrimination of phenotypic subtypes in a reduced dimensional space obtained by Fisher discriminant analysis (FDA). This approach was tested by applying the algorithm to existing data sets for estimation of the minimum sample size required for drawing certain conclusions on multi-class distinction with statistical reliability. It was confirmed that when the minimum number of samples estimated from power analysis is used, group mean in the FDA discrimination space are statistically different.

Original languageEnglish (US)
Pages (from-to)1184-1193
Number of pages10
JournalBioinformatics
Volume18
Issue number9
DOIs
StatePublished - Jan 1 2002
Externally publishedYes

Fingerprint

Discriminant analysis
Microarrays
Microarray Data
Sample Size
Power Analysis
Discriminant Analysis
Tissue
Fisher Discriminant Analysis
Discrimination
Multi-class
Microarray
Availability
Jurisprudence
Profiling
Phenotype
Distinct
Calculate
Estimate

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Determination of minimum sample size and discriminatory expression patterns in microarray data. / Hwang, Daehee; Schmitt, William A.; Stephanopoulos, George; Stephanopoulos, Gregory.

In: Bioinformatics, Vol. 18, No. 9, 01.01.2002, p. 1184-1193.

Research output: Contribution to journalArticle

Hwang, Daehee ; Schmitt, William A. ; Stephanopoulos, George ; Stephanopoulos, Gregory. / Determination of minimum sample size and discriminatory expression patterns in microarray data. In: Bioinformatics. 2002 ; Vol. 18, No. 9. pp. 1184-1193.
@article{7ee5b9ef0de647a3a06378c12d65c924,
title = "Determination of minimum sample size and discriminatory expression patterns in microarray data",
abstract = "Motivation: Transcriptional profiling using microarrays can reveal important information about cellular and tissue expression phenotypes, but these measurements are costly and time consuming. Additionally, tissue sample availability poses further constraints on the number of arrays that can be analyzed in connection with a particular disease or state of interest. It is therefore important to provide a method for the determination of the minimum number of microarrays required to separate, with statistical reliability, distinct disease states or other physiological differences. Results: Power analysis was applied to estimate the minimum sample size required for two-class and multi-class discrimination. The power analysis algorithm calculates the appropriate sample size for discrimination of phenotypic subtypes in a reduced dimensional space obtained by Fisher discriminant analysis (FDA). This approach was tested by applying the algorithm to existing data sets for estimation of the minimum sample size required for drawing certain conclusions on multi-class distinction with statistical reliability. It was confirmed that when the minimum number of samples estimated from power analysis is used, group mean in the FDA discrimination space are statistically different.",
author = "Daehee Hwang and Schmitt, {William A.} and George Stephanopoulos and Gregory Stephanopoulos",
year = "2002",
month = "1",
day = "1",
doi = "10.1093/bioinformatics/18.9.1184",
language = "English (US)",
volume = "18",
pages = "1184--1193",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "9",

}

TY - JOUR

T1 - Determination of minimum sample size and discriminatory expression patterns in microarray data

AU - Hwang, Daehee

AU - Schmitt, William A.

AU - Stephanopoulos, George

AU - Stephanopoulos, Gregory

PY - 2002/1/1

Y1 - 2002/1/1

N2 - Motivation: Transcriptional profiling using microarrays can reveal important information about cellular and tissue expression phenotypes, but these measurements are costly and time consuming. Additionally, tissue sample availability poses further constraints on the number of arrays that can be analyzed in connection with a particular disease or state of interest. It is therefore important to provide a method for the determination of the minimum number of microarrays required to separate, with statistical reliability, distinct disease states or other physiological differences. Results: Power analysis was applied to estimate the minimum sample size required for two-class and multi-class discrimination. The power analysis algorithm calculates the appropriate sample size for discrimination of phenotypic subtypes in a reduced dimensional space obtained by Fisher discriminant analysis (FDA). This approach was tested by applying the algorithm to existing data sets for estimation of the minimum sample size required for drawing certain conclusions on multi-class distinction with statistical reliability. It was confirmed that when the minimum number of samples estimated from power analysis is used, group mean in the FDA discrimination space are statistically different.

AB - Motivation: Transcriptional profiling using microarrays can reveal important information about cellular and tissue expression phenotypes, but these measurements are costly and time consuming. Additionally, tissue sample availability poses further constraints on the number of arrays that can be analyzed in connection with a particular disease or state of interest. It is therefore important to provide a method for the determination of the minimum number of microarrays required to separate, with statistical reliability, distinct disease states or other physiological differences. Results: Power analysis was applied to estimate the minimum sample size required for two-class and multi-class discrimination. The power analysis algorithm calculates the appropriate sample size for discrimination of phenotypic subtypes in a reduced dimensional space obtained by Fisher discriminant analysis (FDA). This approach was tested by applying the algorithm to existing data sets for estimation of the minimum sample size required for drawing certain conclusions on multi-class distinction with statistical reliability. It was confirmed that when the minimum number of samples estimated from power analysis is used, group mean in the FDA discrimination space are statistically different.

UR - http://www.scopus.com/inward/record.url?scp=0036738906&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036738906&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/18.9.1184

DO - 10.1093/bioinformatics/18.9.1184

M3 - Article

C2 - 12217910

AN - SCOPUS:0036738906

VL - 18

SP - 1184

EP - 1193

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 9

ER -