Microarrays Research Today is a free monthly online journal that collates and summarizes the latest research about Microarrays, including details on experiments, designs, statistics, analysis, software.

Application of reliability coefficients in cDNA microarray data analysis.

He W, Bull SB, Gokgoz N, Andrulis I, Wunder J

Prosserman Center for Health Research, Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X5. whe@stats.uwo.ca

Gene expression microarray technology has been widely used in areas such as human cancer research to identify molecular characteristics of sample specimens. The microarray study, however, is a very complicated procedure which involves numerous sources of variability that may be either systematic or random. Systematic variation is often eliminated by applying normalization procedures, but at present there are no standard criteria available to evaluate the performance of a particular normalization approach. In this paper, we propose a reliability-type coefficient as a criterion to assess the effectiveness of normalization procedures in eliminating systematic variation. Simulation studies show that this criterion performs reasonably well in a range of settings. The proposed method is illustrated using a subset of an ongoing microarray study of soft-tissue sarcoma.

Published 15 February 2006 in Stat Med, 25(6): 1051-66.
Full-text of this article is available online (may require subscription).

© 2004-2008 Microarrays Research Today. All Rights Reserved.