Simultaneously testing multiple hypotheses is important in high-dimensional biological studies. In these situations, one is often interested in controlling the Type-I error rate, such as the proportion of false positives to total rejections (TPPFP) at a specific level, alpha. This article will present an application of the E-Bayes/Bootstrap TPPFP procedure, presented in van der Laan et al. (2005), which controls the tail probability of the proportion of false positives (TPPFP), on two biological datasets. The two data applications include firstly, the application to a mass-spectrometry dataset of two leukemia subtypes, AML and ALL. The protein data measurements include intensity and mass-to-charge (m/z) ratios of bone marrow samples, with two replicates per sample. We apply techniques to preprocess the data; i.e. correct for baseline shift of the data as well as appropriately smooth the intensity profiles over the m/z values. After preprocessing the data we show an application of a TPPFP multiple testing techniques (van der Laan et al. (2005)) to test the difference between two groups of patients (AML/ALL) with respect to their intensity values over various m/z ratios, thus indicative of testing proteins of different sizes. Secondly, we will show an illustration of the E-Bayes/Bootstrap TPPFP procedure on a bacterial data set. In this application we are interested in finding bacteria whose mean difference over time points is differentially expressed between two U.S. cities. With both of these data applications, we also show comparisons to the van der Laan et al. (2004b) tppfp augmentation method, and discover the E-Bayes/Bootstrap TPPFP method is less conservative, therefore rejecting more tests at a specific alpha level
Biostatistics | Genetics | Laboratory and Basic Science Research | Statistical Methodology | Statistical Models | Statistical Theory
Birkner, Merrill D.; Hubbard, Alan E.; and van der Laan, Mark J., "Application of a Multiple Testing Procedure Controlling the Proportion of False Positives to Protein and Bacterial Data" (August 2005). U.C. Berkeley Division of Biostatistics Working Paper Series. Working Paper 186.