CHANGES IN VERSION 1.7.2
------------------------

NEW FEATURES

    o New method 'edivisive' available.
    
    o Breakpoint detection available for both Aneufinder(..., strandseq=TRUE) and Aneufinder(..., strandseq=FALSE)
    
    o A stepsize for a sliding window can be selected in addition to the binsize for method "HMM". This improves resolution of detected breakpoints.
    
    o Breakpoints for Aneufinder(..., strandseq=TRUE) are reported with confidence intervals in folder BROWSERFILES.
    
    o Breakpoint detection for Aneufinder(..., strandseq=TRUE) has an additional breakpoint refinement step which improves localization of breakpoints.

SIGNIFICANT USER-LEVEL CHANGES

    o Reorganized output folder structure and added README.txt

    o Renamed parameter "plot.SCE" to "plot.breakpoints".
    
    o GC correction is now done with a loess-fit by default instead of the quadratic fit. This should improve accuracy.
    
    o Default epsilon is now 'eps=0.01' (was 'eps=0.1' before) for method "HMM".
    
    
CHANGES IN VERSION 1.5.1
------------------------

BUG FIXES

    o Can use BSgenome*NCBI* for GC correction (worked only for BSgenome*UCSC* before).
    
    o Bugfix in bam2GRanges(..., pairedEndReads=TRUE) if both alignments are not on the same chromosome.
    
    
CHANGES IN VERSION 1.3.4
------------------------

NEW FEATURES

    o Proper print() methods for AneuFinder objects.
    
    o plotProfile(..., normalize.counts='2-somy') option added for plotting of normalized counts.
    
    o heatmapGenomewideCluster() added for convenient assessment of the clusterByQuality() result.
    
BUG FIXES

    o Fixed option Aneufinder(..., strandseq = TRUE) for DNAcopy method.
    
    o Fixed a bug for SCE plotting in heatmapGenomewide().
    
    o Proper creation of variable-width bins for huge reference files.
    
    o Better heatmap dimensions for few cells.
    
    o Bugfix for Inf values in clusterByQuality().
    
SIGNIFICANT USER-LEVEL CHANGES

    o Renamed plotPCA() to plot_pca() to avoid namespace conflict with BiocGenerics.
    

CHANGES IN VERSION 1.3.3
------------------------

NEW FEATURES

    o New function plotPCA() to do principal component analysis.

    o Introduced parameter 'exclude.regions' to karyotypeMeasures(), plotHeterogeneity() and heatmapGenomewide(). This should facilitate excluding artifact regions from the clustering and karyotype measures.

    o Parameter 'regions' now also available in plotHeterogeneity() (only in karyotypeMeasures() before).

SIGNIFICANT USER-LEVEL CHANGES

    o The method to compute the dendrogram in heatmapGenomewide() was changed to simple hierarchical clustering on the copy number at bin-level (was segment-level before).


CHANGES IN VERSION 1.3.1
------------------------

NEW FEATURES

    o Parameter 'normalChromosomeNumbers' in karyotypeMeasures() can handle mixture samples now.


CHANGES IN VERSION 1.1.6
------------------------

NEW FEATURES

    o Added DNAcopy algorithm to Strand-seq mode.

SIGNIFICANT USER-LEVEL CHANGES

    o Renamed parameter 'most.frequent.state.bivariate' -> 'most.frequent.state.strandseq' in Aneufinder().

    o Renamed parameter 'most.frequent.state.univariate' -> 'most.frequent.state' in Aneufinder().

    o New parameter 'strandseq' in Aneufinder().

BUG FIXES

    o Dendrogram and heatmap are now aligned properly in heatmapGenomewide().


CHANGES IN VERSION 1.1.5
------------------------

NEW FEATURES

    o Aneufinder runs DNAcopy algorithm in addition to the Hidden Markov Model.

    o New function "getQC" to get a data.frame with quality metrics.

SIGNIFICANT USER-LEVEL CHANGES

    o Changed folder structure to include DNAcopy method.

    o Renamed methods from c('univariate','bivariate') to c('HMM','biHMM')


CHANGES IN VERSION 1.1.4
------------------------

NEW FEATURES

    o karyotypeMeasures() has new option regions.

    o plotHeterogeneity() for easy plotting of karyotype measures.

    o BiocStyle vignette.

    o New option use.bamsignals=FALSE/TRUE available for the binning step.

    o getQC() handles NULL entries as NA and is thus more robust.

    o complexity estimation via Michaelis-Menten is carried along.

SIGNIFICANT USER-LEVEL CHANGES

    o Color scheme for copy number states has been improved for states >= 5-somy.

    o Option format has been removed in all functions. File format is determined
      automatically now.

    o clusterByQuality() clusters now on complexity as well by default.

BUG FIXES

    o Corrected bug in order of seqlevels after as(..., 'GRanges').

    o Corrected bug in hotspotter() that caused detection of low-abundance hotspots.