´Tuyển tập báo cáo các nghiên cứu khoa học quốc tế ngành hóa học dành cho các bạn yêu hóa học tham khảo đề tài: The L1-Version of the Cramer-von Mises Test for Two-Sample Comparisons in Microarray Data Analysis | Hindawi Publishing Corporation EURASIP Journal on Bioinformatics and Systems Biology Volume 2006 Article ID 85769 Pages 1-9 DOI BSB 2006 85769 The LI-Version of the Cramer-von Mises Test for Two-Sample Comparisons in Microarray Data Analysis Yuanhui Xiao 1 2 Alexander Gordon 1 3 and Andrei Yakovlev1 1 Department of Biostatistics and Computational Biology University of Rochester 601 Elmwood Avenue . Box 630 Rochester NY 14642 USA 2 Department of Mathematics and Statistics Georgia State University Atlanta GA 30303 USA 3 Department of Mathematics and Statistics University of North Carolina at Charlotte 9201 University City Boulevard Charlotte NC 28223 USA Received 31 January 2006 Accepted 27 June 2006 Recommended for Publication by Jaakko Astola Distribution-free statistical tests offer clear advantages in situations where the exact unadjusted p-values are required as input for multiple testing procedures. Such situations prevail when testing for differential expression of genes in microarray studies. The Cramer-von Mises two-sample test based on a certain L2-distance between two empirical distribution functions is a distribution-free test that has proven itself as a good choice. A numerical algorithm is available for computing quantiles of the sampling distribution of the Cramer-von Mises test statistic in finite samples. However the computation is very time- and space-consuming. An L1 counterpart of the Cramer-von Mises test represents an appealing alternative. In this work we present an efficient algorithm for computing exact quantiles of the L1-distance test statistic. The performance and power of the L1-distance test are compared with those of the Cramer-von Mises and two other classical tests using both simulated data and a large set of microarray data on childhood leukemia. The L1-distance test appears to be nearly as powerful as its L2 counterpart. The lower computational intensity of the L1 -distance test allows computation of exact quantiles of .
