Updated to version 1.4.1 in the release (13/11/2013)
Updated to version 1.5.3 (10/11/2013)
08/11/2013      fixed the bug in importPublicData function that caused problems when importing only one sample
08/11/2013	fixed error in BibTex file
Updated to version 1.5.2 (06/11/2013)
06/11/2013	updated vignette and NEWS file
Updated to version 1.5.1 (02/11/2013)
02/11/2013	moved vignette source to /vignettes folder
Updated to version 1.5.0
15/10/2013	Biodonductor 2.14. release
Updated to version 1.3.10 in the development (24/09/2013)
24/09/2013	optimized calculation of K-S p-values for shifting promoters using vectorized functions
23/09/2013	removed a .C call to the pKS2 function from stats package and replaced it by an R implementation of pKS2
Updated to version 1.2.9 in the release and 1.3.9 in the development (23/08/2013)
23/08/2013      added consensusClustersTpm() function for retrieving matrix with tpm values for consensus clusters across all samples
Updated to version 1.2.8 in the release and 1.3.8 in the development (23/08/2013)
22/08/2013	modified all functions to work on data from only one strand (some were previously giving error in such case)
21/08/2013	fixed the bug in calling CTSSs from reads on minus strand when not correcting for G addition, i.e. when removeFirstG=FALSE
Updated to version 1.2.7 in the release and 1.3.7 in the development (08/08/2013)
08/08/2013	added new argument to aggregateTagClusters() function (excludeSignalBelowThreshold) that controls which TCs will contribute to the total CAGE signal of the consensus cluster - only the ones above the threshold that are used initially to set the boundaries of the consensus cluster, or all TCs that overlap the resulting consensus clusters 
07/08/2013	enabled retrieving interquantile width for tag clusters - added new returnInterquantileWidth argument to tagClusters() function
06/07/2013	enabled importing entire CTSS table with multiple columns of tag counts representing multiple samples/experiments - allowed setting imputFilesType to "CTSStable" when creating CAGEset object and extended getCTSS function accordingly to read the entire table at once and fill the object
Updated to version 1.2.6 in the release and 1.3.6 in the development (23/07/2013)
22/07/2013	replaced multicore package dependency with parallel package (suggested by Bioconductor core team)
21/07/2013	fixed the bug in using loaded genome when correcting for G nucleotide bias
20/07/2013	fixed plotReverseCumulatives to work with only one sample in the CAGEset
19/07/2013	changed naming of correlation plot files to distinguish CTSS from consensus clusters plot
Updated to version 1.2.5 in the release and 1.3.5 in the development (28/06/2013)
28/06/2013	updated the vignette
Updated to version 1.2.4 in the release and 1.3.4 in the development (16/06/2013)
16/06/2013	allowed plotting reverse cumulatives for normalized tag counts - added new argument to plotReverseCumulatives() function
15/06/2013	implemented calculating suggested referent power-law distribution and visualizing the distribution and its parameters on reverse cumulatives plot when calling plotReverseCumulatives() function
14/06/2013	allowed assigning user-specified colors to samples to be used in visualizations - added new setColors() function
13/06/2013	allowed plotting scatter plots and correlation for both individual TSSs (raw or normalized values) and consensus clusters - added new argument to plotCorrelation() function
Updated to version 1.2.3. in the release and 1.3.3 in the development (12/06/2013)
12/06/2013	implemented plotting pairwise scatter plots of CAGE tag count and calculating correlation between samples - added new plotCorrelation() function
Updated to version 1.2.2. in the release and 1.3.2 in the development (30/05/2013)
30/05/2013	implemented the algorithm for correcting 'G' nucleotide addition bias to CAGE tags described in Carninci et al., Nature Genetics 2006 - added a new option 'correctSystematicG' to getCTSS function
Updated to version 1.2.1 in the release and 1.3.1 in the development (28/05/2013)
28/05/2013	fixed the bug in reporting number of CTSSs in cluster when clustering CTSSs using custom clusters (for empty clusters it was returning that the number of CTSSs is 1, although the signal is 0)
28/05/2013	fixed the bug in clustering CTSSs using custom clusters when run with multicore=FALSE (it was returning to many clusters - correct specified region but all of the regions on each chromosome, not only the specified one)
22/05/2013	fixed the column classes in CTSS data.frame when reading the data from bam files (to match the classes when reading from ctss files)
21/05/2013	updated the vignette with explanation on how to use custom build genomes
21/05/2013	allowed usage of custom build BSgenome packages by removing the check whether the specified genome is present in the available.genomes() from BSgenome package
Updated to version 1.2.0 in the release and 1.3.0 in the development (02/05/2013)
02/05/2013	updated the vignette
02/05/2013	added a new feature in the plotReverseCumulatives function, so that the slope of the fitted power-law distribution in the user selected range of values is shown on the plot for each sample (helps to choose appropriate alpha parameter for normalization)
02/05/2013	fixed a bug in selecting the range of tag count values for fitting power-law distribution in normalization
01/05/2013	added sample labels as names to library sizes vector
29/04/2013	implemented statistical testing (Kolmogorov-Smirnov test) for differential promoter usage based on cumulative sums of CAGE signal along the promoters (implemented within the scoreShift function)
20/04/2013	added pass-through option "none" for 'method' parameter in normalizeTagCount function to enable using raw tag counts in downstream steps (CTSS clustering, promoter width, etc.)
19/04/2013	optimized scoreShift function to extract and process only the cumulative sums for samples being compared 
19/04/2013	replaced lapply with a for loop in the scoreShift function to avoid invoking multicore within lapply
18/04/2013	fixed wrong error message in plotExpressionProfiles function that notifies about accepted values for 'what' parameter
27/03/2013	added data.table to the list of dependencies and optimized various parts of the code to use data.table instead of data.frame
