BasicSTARRseq: a Bioconductor R-package for analyzing STARR-seq data

Self-transcribing active regulatory region sequencing (STARR-seq) was first described in 2013 by Arnold et al. and allows to identify and quantify enhancer regions in non-coding DNA in large scale. The R-package BasicSTARRseq provides routines for quality controls, analysis and visualization of STARR-seq data. The analysis part is mainly covered through the implementation of the computational procedure to call peaks i. e. identify possible enhancers, which was introduced in the above mentioned article. The peak calling is based on comparing sample data with input data of the STARR-seq experiment and computes p-values to estimate the peaks' reliability. By including user chosen parameters, for example two alternative binomial models for calculating the p-value, peak calling can be adjusted to different kinds of data. The procedure can further be adapted to whole-genome or targeted sequencing. Resulting peaks are annotated to allow an easy overview over the results or for further filtering steps. Quality controls and visualization are offered by routines for comparing different replicates, and the comparison of experiment data and target regions. For plausibility checks or further explorative analysis the package also provides some functions to compare output tracks of other analysis (like peak lists of Chip-seq data, but also other data chosen by the user) with STARR-seq data. BacisSTARRseq includes test datasets extracted from the published data of Arnold et al.