ABSTRACT
This chapter explains the transform achieved false discovery rate into local false discovery rate estimates. Conventional methods of adjusting p values for multiple comparisons control a family-wise error rate such as a genome-wise error rate. The recognition that they lead to excessive false negative rates in genomics applications has led to widespread use of false discovery rate in place of the conventional adjustments. Selecting a gene for further consideration on the basis of accurate false discovery rate estimation in effect selects a gene on the basis of the probability that a more significant gene is differentially expressed. That is equivalent to selecting an applicant for admission on the basis of an average of his or her probability of failure with the failure probabilities of all the better applicants.
