How Many Mice and How Many Arrays? Replication in Mouse cDNA Microarray Experiments

Biological and technical variances were estimated from the Project Normal data using the mixed model analysis of variance. The technical variance is larger than the biological variance in most genes. In experiments for detecting treatment effects using a reference design, increasing the number of mice per treatment is more effective than pooling mice or increasing the number of arrays per mouse. For a given number of arrays, more mice per treatment with fewer arrays per mouse are more powerful than fewer mice per treatment with more arrays per mouse. A formula is provided for computing the optimum number of arrays per mouse to minimize the total cost of the experiment.

[1]  P. Nelson,et al.  Project normal: Defining normal variance in mouse gene expression , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[2]  G. A. Whitmore,et al.  Importance of replication in microarray gene expression studies: statistical methods and evidence from repetitive cDNA hybridizations. , 2000, Proceedings of the National Academy of Sciences of the United States of America.

[3]  T. Speed,et al.  Design issues for cDNA microarray experiments , 2002, Nature Reviews Genetics.

[4]  R. Kuehl.  Design of Experiments: Statistical Principles of Research Design and Analysis , 1999 .

[5]  Viktor Witkovsk,et al.  MATLAB algorithm mixed.m for solving Henderson's mixed model equations , 2002 .

[6]  G. Churchill.  Fundamentals of experimental design for cDNA microarrays , 2002, Nature Genetics.

[7]  Pierre Baldi,et al.  A Bayesian framework for the analysis of microarray expression data: regularized t -test and statistical inferences of gene changes , 2001, Bioinform..

[8]  Pierre R. Bushel,et al.  Assessing Gene Significance from cDNA Microarray Expression Data via Mixed Models , 2001, J. Comput. Biol..

[9]  Juliane Fluck,et al.  Microarrays: how many do you need? , 2002, RECOMB '02.

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

[11]  M. Radmacher,et al.  Design of studies using DNA microarrays , 2002, Genetic epidemiology.

[12]  Partha S. Vasisht.  Computational Analysis of Microarray Data , 2003 .

[13]  John D. Storey.  A direct approach to false discovery rates , 2002 .

[14]  Gary A. Churchill,et al.  Analysis of Variance for Gene Expression Microarray Data , 2000, J. Comput. Biol..

[15]  Karl J. Friston,et al.  Variance Components , 2003 .

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

[17]  G. Churchill,et al.  Experimental design for gene expression microarrays. , 2001, Biostatistics.

[18]  X. Cui,et al.  Statistical tests for differential expression in cDNA microarray experiments , 2003, Genome Biology.

[19]  W. Pan,et al.  How many replicates of arrays are required to detect gene expression changes in microarray experiments? A mixture model approach , 2002, Genome Biology.