- trim process will now always output fastp report regardless of
--keep_trim - improve documentation
- input is now a samplesheet indicating sample name and the R1/R2 files plus a group information.
- separated trimming and alignment/sort into separate processes
- allow multiple fastq files per sample, will be merged prior to alignment by a dedicated process
- called peaks are now filtered against provided blacklists, currently mm10 and hg38 are supported
- now publish software versions and command lines (the latter separated by process and sample) into a dedicated pipeline directory
- major code cleanup
- new container versions
- better documentation
- added error strategy
finishto all modules - all modules use the same publishing mode which by default is
copy - some eye-candy in the startup message
- only run with <= 21.04.3 as in later versions include statements must be outside the workflow definitions, which we have not yet done
- remove the conda CI test, only run with Docker
(...)