Adaptive filtering increases power to detect differentially expressed genes

Zixin Nie; Zixin Nie; Kun Liang

Adaptive filtering increases power to detect differentially expressed genes

Nie, Z., & Liang, K. (2017). Adaptive filtering increases power to detect differentially expressed genes. In New Advances in Statistics and Data Science (pp. 127-136). Springer. https://link.springer.com/chapter/10.1007/978-3-319-69416-0_8

Copy citation

Abstract

Detecting differentially expressed genes is difficult due to the large number of genes simultaneously tested, resulting in low power for each test after adjusting for multiplicity. We propose a novel adaptive filtering procedure that improves power by filtering out genes that are unlikely to be differentially expressed. We show that the proposed procedure controls the false discovery rate asymptotically. Simulation study further demonstrate its advantage over the state-of-the-art competitors.