AdaFDR: A Fast, Powerful and Covariate-Adaptive Approach to Multiple Hypothesis Testing

摘要

Multiple hypothesis testing is an essential component in many modern data analysis workflows. A very common objective is to maximize the number of discoveries while controlling the fraction of false discoveries. For example, we may want to identify as many genes as possible that are differentially expressed between two populations such that less than, say, 10% of these identified genes are false positives.