Detecting Differential Expressions in GeneChip Microarray Studies

In this article we consider testing for differentially expressed genes in GeneChip studies by modeling and analyzing the quantiles of gene expression through probe level measurements. By developing a robust rank score test for linear quantile models with a random effect, we propose a reliable test for detecting differences in certain quantiles of the intensity distributions. By using a genomewide adjustment to the test statistic to account for within-array correlation, we demonstrate that the proposed rank score test is highly effective even when the number of arrays is small. Our empirical studies with real experimental data show that detecting differences in the quartiles for the probe level data is a valuable complement to the usual mixed model analysis based on Gaussian likelihood. The methodology proposed in this article is a first attempt to develop inferential tools for quantile regression in mixed models.

[1]  B. Weir,et al.  A systematic statistical linear modeling approach to oligonucleotide array experiments. , 2002, Mathematical biosciences.

[2]  Q. Shao,et al.  A general bahadur representation of M-estimators and its application to linear regression with nonstochastic designs , 1996 .

[3]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[4]  Ingrid Lönnstedt.  Replicated microarray data , 2001 .

[5]  Jianqing Fan,et al.  Semilinear High-Dimensional Model for Normalization of Microarray Data , 2005 .

[6]  R. Koenker,et al.  Regression Quantiles , 2007 .

[7]  Zoltan Szallasi,et al.  Increased measurement accuracy for sequence-verified microarray probes. , 2004, Physiological genomics.

[8]  Zhongyi Zhu,et al.  Estimation in a semiparametric model for longitudinal data with unspecified dependence structure , 2002 .

[9]  C. Li,et al.  Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[10]  Xuming He,et al.  Practical Confidence Intervals for Regression Quantiles , 2005 .

[11]  Rafael A Irizarry,et al.  Exploration, normalization, and summaries of high density oligonucleotide array probe level data. , 2003, Biostatistics.

[12]  Jianhua Hu,et al.  Enhanced Quantile Normalization of Microarray Data to Reduce Loss of Information in Gene Expression Profiles , 2007, Biometrics.

[13]  Kenny Q. Ye.  Statistical Tests for Mixed Linear Models , 2000, Technometrics.

[14]  Jianqing Fan,et al.  Removing intensity effects and identifying significant genes for Affymetrix arrays in macrophage migration inhibitory factor-suppressed neuroblastoma cells. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[15]  Roger Koenker,et al.  L-Estimation for Linear Models , 1987 .

[16]  R. Littell.  SAS System for Mixed Models , 1996 .

[17]  Gang Liu,et al.  Effects of cigarette smoke on the human airway epithelial cell transcriptome. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[18]  Roger Koenker,et al.  Tests of linear hypotheses based on regression rank scores , 1993 .

[19]  R. Koenker.  Quantile regression for longitudinal data , 2004 .