Nested, non-parametric, correlative analysis of microarrays for heterogenous phenotype characterization

Jeanne Kowalski, Amanda Blackford, Changyong Feng, Adam J. Mamelak, Daniel N. Sauder

Research output: Contribution to journalArticle

Abstract

We present a non-parametric approach for qualitatively selecting candidate genes to characterize several criteria that are nested among genes selected on the basis of their individual, similar effects upon an array-wide closeness measure. In this setting, a goal is to obtain a reliable characterization of phenotypes, based on very high-dimensional data from a few samples. As opposed to a distance-based approach, the proposed measure defines closeness based on gene signal profiles (functionals) rather than on isolated (numerical) differences in each gene between samples. By using such a measure to characterize intensity differences, we effectively separate biological from artifactual variation in expression, due to tissue effects or signal calibration. Based on this measure, we successively examine the significance of the following: a set of similarly behaved genes relative to all arrayed genes, a set of candidate genes relative to similarly behaved genes, individual candidate genes relative to non-candidates, and the direction, as over- or under-expressed, of candidate genes. In each setting, sample pairs are the units of analysis, with U-statistics the theoretical framework. We illustrate the method on a microarray experiment, where the goal is to select sets of genes that characterize a type of skin cancer and its histological subtypes.

Original languageEnglish (US)
Pages (from-to)1090-1101
Number of pages12
JournalStatistics in Medicine
Volume26
Issue number5
DOIs
StatePublished - Feb 28 2007

Fingerprint

Microarray Analysis
Microarray
Phenotype
Gene
Genes
Nested Genes
U-statistics
Skin Neoplasms
High-dimensional Data
Calibration
Skin
Cancer
Unit

Keywords

  • High-dimensional
  • Inner product
  • Non-parametric

ASJC Scopus subject areas

  • Epidemiology

Cite this

Nested, non-parametric, correlative analysis of microarrays for heterogenous phenotype characterization. / Kowalski, Jeanne; Blackford, Amanda; Feng, Changyong; Mamelak, Adam J.; Sauder, Daniel N.

In: Statistics in Medicine, Vol. 26, No. 5, 28.02.2007, p. 1090-1101.

Research output: Contribution to journalArticle

Kowalski, Jeanne ; Blackford, Amanda ; Feng, Changyong ; Mamelak, Adam J. ; Sauder, Daniel N. / Nested, non-parametric, correlative analysis of microarrays for heterogenous phenotype characterization. In: Statistics in Medicine. 2007 ; Vol. 26, No. 5. pp. 1090-1101.
@article{ea50ea28b90944f187c407d57c39a0a1,
title = "Nested, non-parametric, correlative analysis of microarrays for heterogenous phenotype characterization",
abstract = "We present a non-parametric approach for qualitatively selecting candidate genes to characterize several criteria that are nested among genes selected on the basis of their individual, similar effects upon an array-wide closeness measure. In this setting, a goal is to obtain a reliable characterization of phenotypes, based on very high-dimensional data from a few samples. As opposed to a distance-based approach, the proposed measure defines closeness based on gene signal profiles (functionals) rather than on isolated (numerical) differences in each gene between samples. By using such a measure to characterize intensity differences, we effectively separate biological from artifactual variation in expression, due to tissue effects or signal calibration. Based on this measure, we successively examine the significance of the following: a set of similarly behaved genes relative to all arrayed genes, a set of candidate genes relative to similarly behaved genes, individual candidate genes relative to non-candidates, and the direction, as over- or under-expressed, of candidate genes. In each setting, sample pairs are the units of analysis, with U-statistics the theoretical framework. We illustrate the method on a microarray experiment, where the goal is to select sets of genes that characterize a type of skin cancer and its histological subtypes.",
keywords = "High-dimensional, Inner product, Non-parametric",
author = "Jeanne Kowalski and Amanda Blackford and Changyong Feng and Mamelak, {Adam J.} and Sauder, {Daniel N.}",
year = "2007",
month = "2",
day = "28",
doi = "10.1002/sim.2596",
language = "English (US)",
volume = "26",
pages = "1090--1101",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "5",

}

TY - JOUR

T1 - Nested, non-parametric, correlative analysis of microarrays for heterogenous phenotype characterization

AU - Kowalski, Jeanne

AU - Blackford, Amanda

AU - Feng, Changyong

AU - Mamelak, Adam J.

AU - Sauder, Daniel N.

PY - 2007/2/28

Y1 - 2007/2/28

N2 - We present a non-parametric approach for qualitatively selecting candidate genes to characterize several criteria that are nested among genes selected on the basis of their individual, similar effects upon an array-wide closeness measure. In this setting, a goal is to obtain a reliable characterization of phenotypes, based on very high-dimensional data from a few samples. As opposed to a distance-based approach, the proposed measure defines closeness based on gene signal profiles (functionals) rather than on isolated (numerical) differences in each gene between samples. By using such a measure to characterize intensity differences, we effectively separate biological from artifactual variation in expression, due to tissue effects or signal calibration. Based on this measure, we successively examine the significance of the following: a set of similarly behaved genes relative to all arrayed genes, a set of candidate genes relative to similarly behaved genes, individual candidate genes relative to non-candidates, and the direction, as over- or under-expressed, of candidate genes. In each setting, sample pairs are the units of analysis, with U-statistics the theoretical framework. We illustrate the method on a microarray experiment, where the goal is to select sets of genes that characterize a type of skin cancer and its histological subtypes.

AB - We present a non-parametric approach for qualitatively selecting candidate genes to characterize several criteria that are nested among genes selected on the basis of their individual, similar effects upon an array-wide closeness measure. In this setting, a goal is to obtain a reliable characterization of phenotypes, based on very high-dimensional data from a few samples. As opposed to a distance-based approach, the proposed measure defines closeness based on gene signal profiles (functionals) rather than on isolated (numerical) differences in each gene between samples. By using such a measure to characterize intensity differences, we effectively separate biological from artifactual variation in expression, due to tissue effects or signal calibration. Based on this measure, we successively examine the significance of the following: a set of similarly behaved genes relative to all arrayed genes, a set of candidate genes relative to similarly behaved genes, individual candidate genes relative to non-candidates, and the direction, as over- or under-expressed, of candidate genes. In each setting, sample pairs are the units of analysis, with U-statistics the theoretical framework. We illustrate the method on a microarray experiment, where the goal is to select sets of genes that characterize a type of skin cancer and its histological subtypes.

KW - High-dimensional

KW - Inner product

KW - Non-parametric

UR - http://www.scopus.com/inward/record.url?scp=33847173275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847173275&partnerID=8YFLogxK

U2 - 10.1002/sim.2596

DO - 10.1002/sim.2596

M3 - Article

C2 - 16817149

AN - SCOPUS:33847173275

VL - 26

SP - 1090

EP - 1101

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 5

ER -