Nonparametric estimation of genewise variance for microarray data

Jianqing Fan, Yang Feng, Yue Niu

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Estimation of genewise variance arises from two important applications in microarray data analysis: selecting significantly differentially expressed genes and validation tests for normalization of microarray data. We approach the problem by introducing a two-way nonparametric model, which is an extension of the famous Neyman-Scott model and is applicable beyond microarray data. The problem itself poses interesting challenges because thenumber of nuisance parameters is proportional to the sample size and it is not obvious how the variance function can be estimated when measurements are correlated. In such a high-dimensional nonparametric problem, we proposed two novel nonparametric estimators for genewise variance function and semiparametric estimators for measurement correlation, via solving a system of nonlinear equations. Their asymptotic normality is established. The finite sample property is demonstrated by simulation studies. The estimators also improve the power of the tests for detecting statistically differentially expressed genes. The methodology is illustrated by the data from microarray quality control (MAQC) project.

Original languageEnglish (US)
Pages (from-to)2723-2750
Number of pages28
JournalAnnals of Statistics
Volume38
Issue number5
DOIs
StatePublished - Oct 2010

Fingerprint

Nonparametric Estimation
Microarray Data
Variance Function
Gene
Microarray Data Analysis
Estimator
System of Nonlinear Equations
Nuisance Parameter
Nonparametric Estimator
Nonparametric Model
Quality Control
Asymptotic Normality
Microarray
Normalization
Sample Size
High-dimensional
Directly proportional
Simulation Study
Methodology
Nonparametric estimation

Keywords

  • Correlation correction
  • Gene selection
  • Genewise variance estimation
  • Local linear regression
  • Nonparametric model
  • Validation test

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Nonparametric estimation of genewise variance for microarray data. / Fan, Jianqing; Feng, Yang; Niu, Yue.

In: Annals of Statistics, Vol. 38, No. 5, 10.2010, p. 2723-2750.

Research output: Contribution to journalArticle

Fan, Jianqing ; Feng, Yang ; Niu, Yue. / Nonparametric estimation of genewise variance for microarray data. In: Annals of Statistics. 2010 ; Vol. 38, No. 5. pp. 2723-2750.
@article{2a9bc502103840d0be94b700127ac33f,
title = "Nonparametric estimation of genewise variance for microarray data",
abstract = "Estimation of genewise variance arises from two important applications in microarray data analysis: selecting significantly differentially expressed genes and validation tests for normalization of microarray data. We approach the problem by introducing a two-way nonparametric model, which is an extension of the famous Neyman-Scott model and is applicable beyond microarray data. The problem itself poses interesting challenges because thenumber of nuisance parameters is proportional to the sample size and it is not obvious how the variance function can be estimated when measurements are correlated. In such a high-dimensional nonparametric problem, we proposed two novel nonparametric estimators for genewise variance function and semiparametric estimators for measurement correlation, via solving a system of nonlinear equations. Their asymptotic normality is established. The finite sample property is demonstrated by simulation studies. The estimators also improve the power of the tests for detecting statistically differentially expressed genes. The methodology is illustrated by the data from microarray quality control (MAQC) project.",
keywords = "Correlation correction, Gene selection, Genewise variance estimation, Local linear regression, Nonparametric model, Validation test",
author = "Jianqing Fan and Yang Feng and Yue Niu",
year = "2010",
month = "10",
doi = "10.1214/10-AOS802",
language = "English (US)",
volume = "38",
pages = "2723--2750",
journal = "Annals of Statistics",
issn = "0090-5364",
publisher = "Institute of Mathematical Statistics",
number = "5",

}

TY - JOUR

T1 - Nonparametric estimation of genewise variance for microarray data

AU - Fan, Jianqing

AU - Feng, Yang

AU - Niu, Yue

PY - 2010/10

Y1 - 2010/10

N2 - Estimation of genewise variance arises from two important applications in microarray data analysis: selecting significantly differentially expressed genes and validation tests for normalization of microarray data. We approach the problem by introducing a two-way nonparametric model, which is an extension of the famous Neyman-Scott model and is applicable beyond microarray data. The problem itself poses interesting challenges because thenumber of nuisance parameters is proportional to the sample size and it is not obvious how the variance function can be estimated when measurements are correlated. In such a high-dimensional nonparametric problem, we proposed two novel nonparametric estimators for genewise variance function and semiparametric estimators for measurement correlation, via solving a system of nonlinear equations. Their asymptotic normality is established. The finite sample property is demonstrated by simulation studies. The estimators also improve the power of the tests for detecting statistically differentially expressed genes. The methodology is illustrated by the data from microarray quality control (MAQC) project.

AB - Estimation of genewise variance arises from two important applications in microarray data analysis: selecting significantly differentially expressed genes and validation tests for normalization of microarray data. We approach the problem by introducing a two-way nonparametric model, which is an extension of the famous Neyman-Scott model and is applicable beyond microarray data. The problem itself poses interesting challenges because thenumber of nuisance parameters is proportional to the sample size and it is not obvious how the variance function can be estimated when measurements are correlated. In such a high-dimensional nonparametric problem, we proposed two novel nonparametric estimators for genewise variance function and semiparametric estimators for measurement correlation, via solving a system of nonlinear equations. Their asymptotic normality is established. The finite sample property is demonstrated by simulation studies. The estimators also improve the power of the tests for detecting statistically differentially expressed genes. The methodology is illustrated by the data from microarray quality control (MAQC) project.

KW - Correlation correction

KW - Gene selection

KW - Genewise variance estimation

KW - Local linear regression

KW - Nonparametric model

KW - Validation test

UR - http://www.scopus.com/inward/record.url?scp=77957591909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957591909&partnerID=8YFLogxK

U2 - 10.1214/10-AOS802

DO - 10.1214/10-AOS802

M3 - Article

AN - SCOPUS:77957591909

VL - 38

SP - 2723

EP - 2750

JO - Annals of Statistics

JF - Annals of Statistics

SN - 0090-5364

IS - 5

ER -