Skip to main content
Fig. 4 | Genome Biology

Fig. 4

From: Controlling bias and inflation in epigenome- and transcriptome-wide association studies using the empirical null distribution

Fig. 4

Histogram of test statistics for TWAS on age (a and b) and smoking status (c and d) performed on two cohorts: LifeLines (LL) and Leiden Longevity Study (LLS). The lines represent the three-component normal mixture fitted as estimated using our Bayesian method. The black line represents the fit of the mixture, the red line the fit of the null component (the empirical null distribution with estimated mean and variance reported). The blue and green lines represent the estimated fits of the alternative components (proportion of positively and negatively associated features)

Back to article page