Premium
Application of reliability coefficients in cDNA microarray data analysis
Author(s) -
He Wenqing,
Bull Shelley B.,
Gokgoz Nalan,
Andrulis Irene,
Wunder Jay
Publication year - 2005
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.2254
Subject(s) - normalization (sociology) , computer science , data mining , microarray analysis techniques , gene chip analysis , reliability (semiconductor) , microarray , statistics , computational biology , bioinformatics , mathematics , biology , gene expression , gene , genetics , power (physics) , physics , quantum mechanics , sociology , anthropology
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. Copyright © 2005 John Wiley & Sons, Ltd.