Statistical Analysis of Data of DNA microarray and Protein array experiments: problems and consultation issues Javier Cabrera Dept of Statistics Rutgers University Abstract DNA and protein microarrays represent the latest challenge in applied statistics and statistical consulting. This area is the fastest growing source of inquiry in the statistical consulting environments, especially at universities and at biomedical and pharmaceutical research centers. At the same time our community is relatively unprepared to deal with this new challenge and needs to adopt new strategies that our consulting center models to be successful with the new questions. More specifically I will start with an overview of the questions and the statistical methods that are available. I will also discuss issues related to consulting environments such us: (i) The interdisciplinary nature of the problems and the need to educate the scientists about the statistical methods they will use for their analysis. (ii) The need to provide education to the statisticians in the areas of genetics and biochemistry. (iii) The nature of the new data requires advanced computational resources. Most microarray experiments produce around ten thousand variables and potentially they will produce hundreds of thousands of variables. (iv) In order to cope with the large number of projects in this area I propose that the centers develop web base data analysis procedures that will be used by the scientists to do the basic analysis of their data and consult the statisticians in the more complex issues. This creates the need for more computational expertise and support for the statistical consulting center. Finally I will use this opportunity to introduce a new book that I coauthor with D. Amarartunga from J&J Pharmaceutical Research and Development with the title: “Exploration and Analysis of DNA Microarray and Protein Array Data” J Wiley & Sons June 2003.