Skip to main content
Fig. 2 | Genome Biology

Fig. 2

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

Fig. 2

The genomic inflation factor overestimates inflation in the presence of a moderate proportion of true associations. The box-plot summarizes the estimated inflation for simulated data with different amounts of true associations. One hundred sets of test statistics were generated with different amounts of true associations (20%, 10% and 5%) but without any true inflation; i.e., the inflation factor should be equal to one. The genomic inflation factor was calculated using \(\lambda _{{\chi _{1}^{2}}}\) [9]. A clear dependence on the number of true associations is seen for the genomic inflation factor

Back to article page

Genome Biology

ISSN: 1474-760X

Contact us