Stepped Linear Regression to Accurately Assess Statistical Significance in Batch Confounded Differential Expression Analysis
Batch effects in microarray experiments may lead to systematic shift in expression measurements from one batch to another. It poses great challenge if batches are confounded with the biological groups of interest especially in the estimation of statistical significance, FDR. Even the widely used well-tailored methods such as SAM are not immune to the effects of batch confounding of groups. We propose a stepped linear regression (SLR) method in the context of SAM to re-estimate the expected statistics and FDR in two class analysis to nullify batch effects and get really significant genes. SLR is equally applicable to the other similar methods and multi-group differential expression analysis.
KeywordsDifferential expression SAM Microarray Batch effect
Unable to display preview. Download preview PDF.
- 1.Fare, T.L., Coffey, E.M., Dai, H., He, Y.D., Kessler, D.A., Kilian, K.A., Koch, J.E., LeProust, E., Marton, M.J., Meyer, M.R., Stoughton, R.B., Tokiwa, G.Y., Wang, Y.: Effects of atmospheric ozone on microarray data quality. Analytical Chemistry 75, 4672–4675 (2003)Google Scholar
- 2.Tusher, V.G., Tibshirani, R., Chu, G.: Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. USA. 98, 5116–5121 (2001)Google Scholar
- 3.Chu G., Narasimhan B., Tibshirani R., Tusher V.G.: SAM, significance Analysis of Microarrays, Users guide and technical document Google Scholar
- 11.Efron, B., Tibshirani, R.: On testing the significance of sets of genes, Tech report. Stanford University (August 2006), http://www-stat.stanford.edu/~tibs/ftp/GSA.pdf