©1996-2019 All Rights Reserved. Online Journal of Bioinformatics . You may not store these pages in any form except for your own personal use. All other usage or distribution is illegal under international copyright treaties. Permission to use any of these pages in any other way besides the  before mentioned must be gained in writing from the publisher. This article is exclusively copyrighted in its entirety to OJB publications. This article may be copied once but may not be, reproduced or  re-transmitted without the express permission of the editors. This journal satisfies the refereeing requirements (DEST) for the Higher Education Research Data Collection (Australia). Linking:To link to this page or any pages linking to this page you must link directly to this page only here rather than put up your own page.



Online Journal of Bioinformatics ©


 Volume 7 (2):90-100, 2006.

A computer-intensive method to analyse microarray experiments with a very

small number of replications in case of heteroscedasticity



Muhammad MA1, Neuhäuser M1, 2


1: Institute for Medical Informatics, Biometry and Epidemiology, University Hospital Essen, Hufelandstr. 55, D-45122 Essen and 2: Department of Mathematics and Technique, RheinAhrCampus, Koblenz University of Applied Science, Südallee 2, D-53424 Remagen, Germany.




Muhammad MA, Neuhäuser M, A computer-intensive method to analyse microarray experiments with a very small number of replications in case of heteroscedasticity, Online J Bioinformatics 7 (2):90-100, 2006. In this paper different tests for the nonparametric Behrens-Fisher problem are investigated, i.e. detecting differences in mean gene expression levels is of primary interest; the tests should not be sensitive for possible pure differences in variability. Beside the rank Welch test and the Brunner-Munzel test a modification of Cliff’s method is studied. In contrast to the original method, the proposed modification has an advantage in case of a very small number of replications (sample size). All tests are performed as bootstrap tests and are applied to both real microarray data as well as simulated data. It is demonstrated that the tests can be applied in case of sample sizes as small as five per group. The results indicate that the rank Welch test is the preferred test.


Keywords:    Identification of differentially expressed genes; Bootstrap test; Rank Welch test; Brunner-Munzel test, Cliff’s method