A new statistic for identifying batch effects in high-throughput genomic data that uses guided principal component analysis

Sarah E. Reese, Kellie J. Archer, Terry M Therneau, Elizabeth J. Atkinson, Celine M Vachon, Mariza De Andrade, Jean-Pierre Kocher, Jeanette E Eckel-Passow

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

Motivation: Batch effects are due to probe-specific systematic variation between groups of samples (batches) resulting from experimental features that are not of biological interest. Principal component analysis (PCA) is commonly used as a visual tool to determine whether batch effects exist after applying a global normalization method. However, PCA yields linear combinations of the variables that contribute maximum variance and thus will not necessarily detect batch effects if they are not the largest source of variability in the data. Results: We present an extension of PCA to quantify the existence of batch effects, called guided PCA (gPCA). We describe a test statistic that uses gPCA to test whether a batch effect exists. We apply our proposed test statistic derived using gPCA to simulated data and to two copy number variation case studies: the first study consisted of 614 samples from a breast cancer family study using Illumina Human 660 bead-chip arrays, whereas the second case study consisted of 703 samples from a family blood pressure study that used Affymetrix SNP Array 6.0. We demonstrate that our statistic has good statistical properties and is able to identify significant batch effects in two copy number variation case studies. Conclusion: We developed a new statistic that uses gPCA to identify whether batch effects exist in high-throughput genomic data. Although our examples pertain to copy number data, gPCA is general and can be used on other data types as well.

Original languageEnglish (US)
Pages (from-to)2877-2883
Number of pages7
JournalBioinformatics
Volume29
Issue number22
DOIs
StatePublished - Nov 15 2013

Fingerprint

Passive Cutaneous Anaphylaxis
Principal Component Analysis
Principal component analysis
High Throughput
Batch
Statistic
Genomics
Throughput
Statistics
Blood pressure
Test Statistic
Single Nucleotide Polymorphism
Breast Neoplasms
Blood Pressure
Breast Cancer
Statistical property
Normalization
Linear Combination
Quantify
Chip

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computational Mathematics
  • Statistics and Probability
  • Medicine(all)

Cite this

A new statistic for identifying batch effects in high-throughput genomic data that uses guided principal component analysis. / Reese, Sarah E.; Archer, Kellie J.; Therneau, Terry M; Atkinson, Elizabeth J.; Vachon, Celine M; De Andrade, Mariza; Kocher, Jean-Pierre; Eckel-Passow, Jeanette E.

In: Bioinformatics, Vol. 29, No. 22, 15.11.2013, p. 2877-2883.

Research output: Contribution to journalArticle

@article{251a2f59714840baa1e62d0204ca614b,
title = "A new statistic for identifying batch effects in high-throughput genomic data that uses guided principal component analysis",
abstract = "Motivation: Batch effects are due to probe-specific systematic variation between groups of samples (batches) resulting from experimental features that are not of biological interest. Principal component analysis (PCA) is commonly used as a visual tool to determine whether batch effects exist after applying a global normalization method. However, PCA yields linear combinations of the variables that contribute maximum variance and thus will not necessarily detect batch effects if they are not the largest source of variability in the data. Results: We present an extension of PCA to quantify the existence of batch effects, called guided PCA (gPCA). We describe a test statistic that uses gPCA to test whether a batch effect exists. We apply our proposed test statistic derived using gPCA to simulated data and to two copy number variation case studies: the first study consisted of 614 samples from a breast cancer family study using Illumina Human 660 bead-chip arrays, whereas the second case study consisted of 703 samples from a family blood pressure study that used Affymetrix SNP Array 6.0. We demonstrate that our statistic has good statistical properties and is able to identify significant batch effects in two copy number variation case studies. Conclusion: We developed a new statistic that uses gPCA to identify whether batch effects exist in high-throughput genomic data. Although our examples pertain to copy number data, gPCA is general and can be used on other data types as well.",
author = "Reese, {Sarah E.} and Archer, {Kellie J.} and Therneau, {Terry M} and Atkinson, {Elizabeth J.} and Vachon, {Celine M} and {De Andrade}, Mariza and Jean-Pierre Kocher and Eckel-Passow, {Jeanette E}",
year = "2013",
month = "11",
day = "15",
doi = "10.1093/bioinformatics/btt480",
language = "English (US)",
volume = "29",
pages = "2877--2883",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "22",

}

TY - JOUR

T1 - A new statistic for identifying batch effects in high-throughput genomic data that uses guided principal component analysis

AU - Reese, Sarah E.

AU - Archer, Kellie J.

AU - Therneau, Terry M

AU - Atkinson, Elizabeth J.

AU - Vachon, Celine M

AU - De Andrade, Mariza

AU - Kocher, Jean-Pierre

AU - Eckel-Passow, Jeanette E

PY - 2013/11/15

Y1 - 2013/11/15

N2 - Motivation: Batch effects are due to probe-specific systematic variation between groups of samples (batches) resulting from experimental features that are not of biological interest. Principal component analysis (PCA) is commonly used as a visual tool to determine whether batch effects exist after applying a global normalization method. However, PCA yields linear combinations of the variables that contribute maximum variance and thus will not necessarily detect batch effects if they are not the largest source of variability in the data. Results: We present an extension of PCA to quantify the existence of batch effects, called guided PCA (gPCA). We describe a test statistic that uses gPCA to test whether a batch effect exists. We apply our proposed test statistic derived using gPCA to simulated data and to two copy number variation case studies: the first study consisted of 614 samples from a breast cancer family study using Illumina Human 660 bead-chip arrays, whereas the second case study consisted of 703 samples from a family blood pressure study that used Affymetrix SNP Array 6.0. We demonstrate that our statistic has good statistical properties and is able to identify significant batch effects in two copy number variation case studies. Conclusion: We developed a new statistic that uses gPCA to identify whether batch effects exist in high-throughput genomic data. Although our examples pertain to copy number data, gPCA is general and can be used on other data types as well.

AB - Motivation: Batch effects are due to probe-specific systematic variation between groups of samples (batches) resulting from experimental features that are not of biological interest. Principal component analysis (PCA) is commonly used as a visual tool to determine whether batch effects exist after applying a global normalization method. However, PCA yields linear combinations of the variables that contribute maximum variance and thus will not necessarily detect batch effects if they are not the largest source of variability in the data. Results: We present an extension of PCA to quantify the existence of batch effects, called guided PCA (gPCA). We describe a test statistic that uses gPCA to test whether a batch effect exists. We apply our proposed test statistic derived using gPCA to simulated data and to two copy number variation case studies: the first study consisted of 614 samples from a breast cancer family study using Illumina Human 660 bead-chip arrays, whereas the second case study consisted of 703 samples from a family blood pressure study that used Affymetrix SNP Array 6.0. We demonstrate that our statistic has good statistical properties and is able to identify significant batch effects in two copy number variation case studies. Conclusion: We developed a new statistic that uses gPCA to identify whether batch effects exist in high-throughput genomic data. Although our examples pertain to copy number data, gPCA is general and can be used on other data types as well.

UR - http://www.scopus.com/inward/record.url?scp=84890087087&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890087087&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btt480

DO - 10.1093/bioinformatics/btt480

M3 - Article

C2 - 23958724

AN - SCOPUS:84890087087

VL - 29

SP - 2877

EP - 2883

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 22

ER -