GENE SELECTION WITH JOHNSON’S DISTRIBUTION

Abstract

Microarrays have become increasingly common in biological and medical research.
A major goal of microarray experiments is to determine which genes are differen-
tially expressed between samples. A new approach using the Johnson’s system of
distributions is proposed in this paper to make simultaneous inference concerning
which genes are differentially expressed, in which no specific parametric distribu-
tion is assumed for the gene expression levels. The simulation study shows that
the new approach has better results compared to many existing methods.