New statistical method that helps improve reproducibility of omics research

BBMRI-NL publication in Genome Biology, January 27th, 2017

The scientific publication output from the BBMRI-NL 2.0 project is growing. On January 27, an article (van Iterson et al, 2017) was published in Genome Biology describing a novel Bayesian method to detect and correct for bias and inflation in epigenome- and transcriptome-wide association studies. The method has the crucial characteristic that it is largely independent of the fraction of true associations in the data. It outperforms existing methods by taking advantage of prior knowledge of the distribution and the composition of test statistics.

Current practice is bound to introduce false positive findings. The newly developed method provides a realistic estimate of inflation that does not suffer from a high variability. Moreover, it is the first method to address the previously unrecognized issue of bias in test statistics and optimally reduces the number of false positive findings while preserving statistical power. It can be seamlessly incorporated into existing work-flows for the analysis of EWAS, TWAS, and other omics data.

The method is developed within the framework of BBMRI-NL 2.0, which aims to develop a research infrastructure that can help understand the pathway from genes to disease development

Samples were contributed by LifeLines (, the Leiden Longevity Study(, the Netherlands Twin Registry (, the Rotterdam Study (, and the CODAM study ( All analyses were carried out on the Dutch national e-infrastructure with the support of SURF Cooperative.


Do you want to start working with the new statistical method? It is implemented as Bioconductor package, called BACON.
Link to package:

Link to publication:

Open Access – This article is distributed under the terms of the Creative Commons Attribution 4.0 International License