In this chapter, we discuss a method of selecting differentially expressed genes based on a newly discovered structure termed as the δ-sequence. Together with the nonparametric empirical Bayes methodology, it leads to dramatic gains in terms of the mean numbers of true and false discoveries, and in the stability of the results of testing. Furthermore, its outcomes are entirely free from the log-additive array-specific technical noise. The new paradigm offers considerable scope for future developments in this area of methodological research.