Journal of Advanced Statistics

Download PDF (855.6 KB) PP. 96 - 108 Pub. Date: June 13, 2017

DOI: 10.22606/jas.2017.22003

• Dallas Joder
  Data Science Division, InferenSys Inc., Virginia, United States
• Nusrat Jahan
  
  Department of Mathematics and Statistics, James Madison University, Virginia, United States

An abundance of data from microarray studies are available in publicly-accessible databases. Most of these studies are conducted by university based research labs. It is not uncommon for such studies to run only three or four replicates for each experimental condition tested. With this low sample size and the high variability and multiple testing problems inherent to microarray technology, it is difficult to draw statistically significant conclusions from any one such study. Meta-analysis could improve this situation by combining evidence from related studies to increase statistical power. In this work we discussed several meta-analysis methods for small sample gene expression studies. We compared the performances of the traditional Fisher’s log-sum and Stouffer’s- Z meta-analysis methods, as well as three weighted variants of Stouffer’s method. Higher false discovery rates were observed for the traditional methods compared to the weighted methods.

Weighted meta-analysis, microarray, meta-p value, false discovery rate, integration driven discoveries, integration driven revisions, Salmonella.

[1] R. Sasik, C. Woelk, and J. Corbiel, “Microarray truths and consequences.” Journal of Molecular Endocrinolgy, vol. 33, pp. 1–9, 2004.

[2] D. Rhodes, T. Barrette, M. Rubin, D. Ghosh, and A. Chinnaiyan, “Meta-analysis of microarrays: interstudy validation of gene expression profiles reveals pathway dysregulation in prostate cancer.” Cancer Research, vol. 62, pp. 4427–4433, 2002.

[3] J. Choi, U. Yu, S. Kim, and O. Yoo, “Combining multiple microarray studies and modeling interstudy variation.” Bioinformatics, vol. 19, pp. 84–90, 2003.

[4] i. G. Parmigian, E. Garrett-Mayer, R. Anbazhagan, and E. Gabrielson, “A cross-study comparison of gene expression studies for the molecular classification of lung cancer.” Clinical Cancer Research, vol. 10, pp. 2922–2927, 2004.

[5] J. Stevens and R. Doerge, “Combining multiple microarray studies and modeling interstudy variation.” BMC Bioinformatics, vol. 6, pp. 1–19, 2006.

[6] P. Hu, C. Greenwood, and J. Beyene, “Combining multiple microarray studies and modeling interstudy variation.” Information Systems Frontiers, vol. 8, pp. 9–20, 2006.

[7] J. Storey and R. Tibshirani, “Combining multiple microarray studies and modeling interstudy variation.” Proceedings of the National Academy of Sciences, vol. 100, pp. 9440–9445, 2003.

[8] N. Arricau, D. Hermant, H. Waxin, and M. Popoff, “Molecular characterization of the salmonella typhi stpa protein that is related to both yersinia yope cytotoxin and yoph tyrosine phosphatase.” Research in Microbiology, vol. 148, pp. 21–26, 1997.

[9] W. Kuo, T. Jenssen, A. Butte, L. Ohno-Machado, and I. Kohane, “Analysis of matched mrna measurements from two different microarray technologies.” Bioinformatics, vol. 18, pp. 405–412, 2002.

[10] D. Ghosh, T. Barrette, D. Rhodes, and A. Chinnaiyan, “Analysis of matched mrna measurements from two different microarray technologies.” Functional & Integrative Genomics, vol. 3, pp. 180–188, 2003.

[11] Y. Huang, H. Xu, V. Calian, and J. Hsu, “To permute or not to permute.” Bioinformatics, vol. 22, pp. 2244–2248, 2006.

[12] T. Roy, “The effect of heteroscedasticity and outliers on the permutation t-test.” ournal of Statistical Computation and Simulation, vol. 72, pp. 23–26, 2002.

[13] V. Tusher, R. Tibshirani, and G. Chu, “Significance analysis of microarrays applied to the ionizing radiation response.” Proceedings of the National Academy of Sciences, vol. 98, pp. 5116–5121, 2001.

[14] G. Smyth, “Linear models and empirical bayes methods for assessing differential expression in microarray experiments.” Statistical Applications in Genetics and Molecular Biology, vol. 3, pp. 1–25, 2004.

[15] R. Wetzels, D. Matzke, M. Lee, J. Rouder, J. Iverson, and E. Wagenmakerset, “Statistical evidence in experimental psychology: An empirical comparison using 855 t tests.” Perspectives on Psychological Science, vol. 6, pp. 291–298, 2011.

[16] V. Johnson, “Revised standards for statistical evidence.” Proceedings of the National Academy of Sciences, vol. 110, pp. 19 313–19 317, 2013.

[17] F. Mosteller and R. Fisher, “Questions and answers.” The American Statistician, vol. 2, pp. 30–31, 1948.

[18] S. Stouffer, “Questions and answers.” The American Soldier, vol. 1, 1949.

[19] R. Rosenthal, “Meta-analysis: a review.” Psychosomatic Medicine, vol. 53, pp. 247–271, 1991.

[20] L. Hedges, H. Cooper, and B. Bushman, “Testing the null hypothesis in meta-analysis: a comparison of combined probability and confidence interval procedures.” Psychological Bulletin, vol. 111, pp. 188–194, 1992.

[21] P. Cahan, “Meta-analysis of microarray results: Challenges, opportunities, and recommendations for standardization.” Gene, vol. 401, pp. 12–18, 2007.

[22] G. Alves and Y. Yu, “Accuracy evaluation of the unified p-value from combining correlated p-values.” PLoS One, vol. 3, 2014.

[23] S. Lucchini, “The h-ns-like protein stpa represses the rpos (sigma 38) regulon during exponential growth of salmonella typhimurium.” Molecular Microbiology, vol. 74, pp. 1169–1186, 2009.

[24] B. Bolstad, R. Irizarry, M. Astrand, and T. Speed, “A comparison of normalization methods for high density oligonucleotide array data based on variance and bias.” Bioinformatics, vol. 19, pp. 185–193, 2003.

[25] S. Pyne, B. Futcher, and S. Skiena, “Meta-analysis based on control of false discovery rate: combining yeast chip-chip datasets.” Bioinformatics, vol. 22, pp. 2516–2522, 2006.