2009-Diaz_et_al - Model selection in a global analysis of a...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
C. Díaz, N. Moreno-Sánchez, J. Rueda, A. Reverter, Y. H. Wang and M. J. Carabaño Model selection in a global analysis of a microarray experiment doi: 10.2527/jas.2007-0713 originally published online Oct 10, 2008; 2009.87:88-98. J Anim Sci http://jas.fass.org/cgi/content/full/87/1/88 the World Wide Web at: The online version of this article, along with updated information and services, is located on www.asas.org by on October 22, 2010. jas.fass.org Downloaded from
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
ABSTRACT: Analysis of data from complementary DNA microarray experiments is an area of intense re- search. Options include models at the gene level or at the global level, the latter combining information from all of the profiled genes. In general, a joint analysis is expected to be more powerful than gene-specific analy- ses. Global analysis of microarray data requires fitting a model that jointly performs data normalization and analyses. The objective of this study was to assess the optimality of alternative models for data normalization and analysis in an experiment to identify differentially expressed genes between 2 muscles in Avileña Negra Ibérica calves. Three major groups of models were ex- plored according to several aspects including spatial ar- rangement of spots, other technical sources of variation such as dye effects, assumptions related to effects in- cluded in the model, and gene-specific effects. In addi- tion, 3 sources of heterogeneity of residual variance were investigated. All models were compared by Bayes fac- tors and cross-validation predictive densities. The mod- el that included array-block, dye, muscle, and array-dye as systematic effects and all gene-related components as random effects was the best model for normalization and analysis of these data under heterogeneity of resid- ual variances. Furthermore, level of intensity seemed to be the major source of heteroscedasticity for all mod- els investigated. Such models rendered the best good- ness of fit without compromising the predictive ability. The best model also provided the best performance to detect genes differentially expressed with the lowest false discovery rate. The large differences found for the model comparison criteria across models indicate the importance of defining the factors that more accurately account for experiment-wide variability to ensure cor- rect inference on differential expression of genes. Our results also illustrate the importance of the experimen- tal setup to account for possible sources of bias in the detection of differentially expressed genes. Key words: Bayesian mixed linear model, differential gene expression, false discovery rate, microarray, model selection, normalization ©2009 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2009. 87:88–98 doi:10.2527/jas.2007-0713 INTRODUCTION Microarray data are notoriously noisy. Thus, the aim is to identify which parts of the measured transcript values are due to biological variation (Tu et al., 2002).
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 13

2009-Diaz_et_al - Model selection in a global analysis of a...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online