WeiYa's Work Yard

A dog, who fell into the ocean of statistics, tries to write down his ideas and notes to save himself.

Joint Local False Discovery Rate in GWAS

Posted on (Update: )
Tags: False Discovery Rate, Genome-wide Association Studies

This note is for Jiang, W., & Yu, W. (2017). Controlling the joint local false discovery rate is more powerful than meta-analysis methods in joint analysis of summary statistics from multiple genome-wide association studies. Bioinformatics, 33(4), 500–507.

\[\newcommand\Jlfdr{\mathrm{Jlfdr}} \newcommand\Fdr{\mathrm{Fdr}}\]

propose a novel summary-statistics-based joint analysis method based on controlling the joint local false discovery rate (Jlfdr).

  • prove that the method is the most powerful summary-statistics-based joint analysis method when controlling the false discovery rate at a certain level
  • the Jlfdr-based method achieves higher power than commonly used meta-analysis methods when analyzing heterogeneous datasets from multiple GWASs

two kinds of joint analysis methods:

  • individual-level
  • summary-statistics-based

Jlfdr generalizes the concept of the local false discovery rate (Jlfdr) from the analysis of single study to the joint analysis of multiple studies


Jlfdr and optimal rejection region

\[\Jlfdr(z) = P(H_0\mid \bfz)\]

Fdr is the expectation of Jlfdr, given that the test statistic vector is in the rejection region $R$.

\[\Fdr(R) = E(\Jlfdr(z)\mid z\in R)\]

Implementation of Jlfdr-based method under the Gaussian mixture model

Published in categories Note