Changes in version 1.1.12                        

New features

  - smoothDeviationsKNN(): adaptive EB smoothing of single-cell
    deviations across nearest-neighbor cells.

Bug fixes

  - Fixed wrong denominator in computeDeviationsWeighted
  - Fixed mistake in background variance (affecting the scale of the
    z-scores): it is now the variance of the counts of each bin's
    background peaks (as would be drawn by sampleBackgroundPeaks()), and
    shrinkage is only applied to the expectation.

Changes

  - The z-scores (and things based on it, e.g. motif variability) now
    include a finite-population correction (fpc=TRUE) of the null
    variance.
  - functions now also return an effective N per motif that is used in
    normalizeDevsForSize for more accurate adjustment.

                        Changes in version 1.1.9                        

  - Fixed bug that some SE-like objects were not recognized as such

                        Changes in version 1.0.8                        

  - Added methods for kNN-based background

                        Changes in version 1.0.2                        

Bug fixes

  - Fixed namespace error in getBackgroundBins when using dgCMatrix
    counts
  - Fixed bug in the binning when using frag length bias
  - Fixed delay in betterChromVAR

                        Changes in version 1.0.0                        

Initial release.