A Powerful Discovery Procedure for Large-Scale Significance Testing, with Application to Comparative Microarray Experiments in Response to Different Biomaterials

Abstract:

Several biomaterials have been widely used in the treatment of cancer. However, how these biomaterials alter gene expression is poorly understood. The problem of identifying genes that are differentially expressed across varying biological conditions or in response to different biomaterials based on microarray data is a typical multiple testing problem. In this paper, we focus on FDR control for large-scale multiple testing problems, by our proposed statistics and resampling method, a powerful FDR controlling procedure for large-scale multiple testing problems is provided. Simulations show that, our Fiducial estimator is accurate and stable than other five traditional methods, with satisfactory FDR control. In particular, we propose a generally applicable estimate of the proposed procedure for identifying differentially expressed genes in microarray experiments. This microarray method consistently shows favorable performance over the existing methods. For example, in testing for differential expression between two breast cancer tumor types, the proposed procedure provides increases from 37% to 127% in the number of genes called significant at a false discovery rate of 3%.

• Periodical: Advanced Materials Research (Volumes 311-313)
• Pages: 1661-1666
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.311-313.1661
• Online since: August 2011
• Authors: Pei Jin, Jian Zhang
• Keywords: Bootstrap Method, FDR, Microarray Experiments, Multiple Testing

References

[1] Y. Benjamini and Y. Hochberg: Journal of the Royal Statistical Society: Series B (Statistical Methodology) Vol. 57 (1995), p.289.

[2] Y. Benjamini and W. Liu: Journal of Statistical Planning and Inference Vol. 82 (1999), p.163.

[3] Y. Benjamini and Y. Hochberg: J. Behav. Educ. Statist. Vol. 25 (2000), p.60.

[4] Y. Benjamini and D. Yekutieli: Annals of Statistics Vol. 29 (2001), p.1165.

[5] Y. Benjamini, A. Krieger, and D. Yekutieli: Biometrica Vol. 93 (2006), p.491.

[6] X. Cui and G.A. Churchill: Genome Biology Vol. 4 (2003), p.210.

[7] B. Efron, R.J. Tibshirani, J.D. Storey and V. Tusher: Journal of the American Statistical Association Vol. 96 (2001), p.1151.

[8] I. Hedenfalk , D. Duggan, Y.D. Chen, M. Radmacher, M. Bittner, R. Simon, P. Meltzer, B. Gusterson, M. Esteller, O.P. Kallioniemi and Others: New England Journal of Medicine Vol. 344 (2001), p.539.

DOI: 10.1056/nejm200102223440801

[9] X. Li, X. Xu, and G. Li: Science in China, Series A:Mathematics Vol. 50 (2007), p.75.

[10] J.D. Storey: Journal of the Royal Statistical Society: Series B (Statistical Methodology) Vol. 64 (2002), p.479.

[11] P. Sebastiani, E. Gussoni, I.S. Kohane and M.F. Ramoni: Statistical Science Vol. 18 (2003), p.33.

[12] J.D. Storey, W. Xiao, J.T. Leek, R.G. Tompkins, and R.W. Davis: Proceedings of the National Academy of Sciences Vol. 102(2005), p.12837.

[13] K.W. Tsui and S. Weerahandi: Journal of the American Statistical Association Vol. 84(1989), p.602.

[14] W.Y.S. Wang, B.J. Barratt, D.G. Clayton and J.A. Todd: Nature Reviews Genetics Vol. 6(2005), p.109.

[15] Y. Gavrilov, Y. Benjamini and S.K. Sarkar: Annals of Statistics Vol. 37 (2009), p.619.