Skip to content

Releases: mikewolfe/ChIPseq_pipeline

Support for multiple contigs and summary calculations

26 May 15:31
3e73f7a

Choose a tag to compare

Added

  • Full support for genomes with multiple contigs/chromosomes (#5)
  • Full support of masking regions of the genome from analysis (#11)
  • Full support for normalization based on a known background region (#6)
  • Calculate summaries over regions or locations (#14)
  • Ability to specify parameter strings for certain rules either as a single
    value for every sample or through specifying a column in the metadata sheet
    for by sample control
  • Ability to add pseudocount before within sample normalization. Only implemented for median normalization
  • Support for normalization based on averages over a given set of regions.
    Enables Occupancy Apparent calculations.
  • Support for normalization for occupancy apparent calculations by dynamically
    determining the highest and lowest N genes by average signal
  • Support for reciprocal ratio calculations to keep ratios in a linear scale.
    Inverts all ratios less than one and takes negative value. Then subtracts or
    adds 1 to recenter to zero.
  • Added a specification to exclude Infs and NaNs from bigwig files

Changed

  • Config syntax change for specifying genome inputs
  • Removed general pseudocount specification. Never adds a pseudocount for ratios
  • Substantially updated the bwtools.py module to enable summary calculations
  • Config syntax change for coverage and normalization specification

Bug fixes

  • Fixed an issue when running coverage_and_norm module only (#3)
  • Fixed an issue with the tbb version being too high on some systems causing bowtie2 to fail to run (workflow/envs/alignment.yaml file).
  • Fixed an issue where genome size was not properly read as input to Macs2 in workflow/rules/peak_calling.smk
  • Fixed an issue where filename paths with spaces in them would not input correctly to cutadapt workflow/rules/preprocessing.smk