crupR: a Bioconductor package to predict condition-specific enhancers from ChIP-seq experiments

Author(s): Persia Akbari Omgba,Martin Vingron,Verena Laupert

Affiliation(s): Max Planck Institute for Molecular Genetics



Enhancers are cis-regulatory elements that play a key role in the regulation of gene expression. Epigenetic data is usually exploited to identify them. Their prediction using such data requires a carefully trained classification method. Enhancer activity varies across conditions such as cell types, developmental stages or cancer tissues. Thus, the classification needs to be applicable across datasets of different origins. Further, the allocation of active enhancers to different conditions and the assignment of putative target genes are common objectives when analysing gene regulation. Recently, we presented the workflow CRUP (condition-specific regulatory units prediction) that was designed to address the aforementioned points. However, due to the command-line based implementation of CRUP, downstream analyses still remain time consuming. We now carefully polished and re-engineered CRUP into the Bioconductor package crupR. crupR comprises all steps of the CRUP workflow in a coherent R pipeline and can now be easily combined with common downstream analyses in R. Finally, we changed a crucial step within the workflow to define differential enhancer regions by replacing the permutation test introduced in CRUP with a Kolmogorov–Smirnov test. With this the typical shape of an enhancer region is reflected in a more adequate way which results in a total increase in differential enhancer regions and a decrease in false positive calls. Availability: The R package crupR is available on github via https://github.com/akbariomgba/crupR}{https://github.com/akbariomgba/crupR. Contact: omgba@molgen.mpg.de