Experiments using arrays of cDNA targets to compare patterns of gene expres
sion are beginning to play a prominent role in biogerontology, but drawing
reliable conclusions from the resulting data sets requires careful applicat
ion of statistical methods that discriminate chance events from those likel
y to reflect real differences among the samples under study. This essay dis
cusses flaws in the logic of studies that base their conclusions on ratio c
alculations alone, reviews the multiple comparison traps inherent in high t
hroughput systems that test a very large number of mRNAs simultaneously, an
d advocates a two-stage design in which significance testing applied to exp
loratory data is used to guide a second round of hypothesis-testing experim
ents conducted in a separate set of experimental samples.