We preprocessed the ENCODE DNase-seq catalog (human) and other publicly available data alongside in-house data to provide quick access to a large and diverse tissue repository.

For picking a suitable tissue of interest, search for a cell type that best reflects your cell context of interest. Because DNase-seq data can differ in their quality and, we provide some basic QC metrics for orientation. Per experiment we provide the total number of reads and the number of peaks called (* marks average numbers from merged replicates). Where peak calls were publicly available we used those or MACS2 (https://github.com/taoliu/MACS) peak calls otherwise (default parameters). We also show the number of reads within peaks and the average number of reads per peak (usually more is better). We found the best indication for a good dataset to be the footprint SFR score for a housekeeping transcription factor k-mer. NRF1 shows one of the strongest footprints in every tissue and thus gives a good estimate of expected footprint quality (again roughly more is better).

For publicly available data we also link to the GEO source. Abbreviations for sources are listed under notes. For detailed information about the ENCODE data visit https://www.encodeproject.org. To use your own DNase-seq data, please use our stand alone R-tool. If you would like to share your data please consider releasing the data and notifying us so that we can pre-process it and make the footprints available via this web page.