CHANGES IN VERSION 1.8.8
-------------------------

  o Corrected a typo in a function to get the names from a
  RangedData annotation.

CHANGES IN VERSION 1.8.7
-------------------------

  o Fixed a bug in the summarization when using the 
  "genes", "geneModels" paradigm and annotation in the
  older RangedData format.

CHANGES IN VERSION 1.8.6
-------------------------

  o Fixed an issue where RPKM values were all set to NAs. Thanks to Vijaya 
  Mahalingam Shanmugiah for reporting that issue.

  o Fixed a number of warnings introduced by changes to the underlying IRanges
  package.

CHANGES IN VERSION 1.8.5
-------------------------

  o Added a check for handling cases where BAM files do not report the
  number of alignments through the NH tag. Thanks to Alistair Chalk
  for spotting that one.

CHANGES IN VERSION 1.8.4
-------------------------

  o Added a check to filter bai file from the files gathered by the pattern
  argument.

  o Added a check for multi-mapping reads.

  o Added a deprecation warning for count='genes', summarization='geneModels'.

  o Extended the vignette, with details on the annotation warnings and
  how to resolve them. Added an FAQ section.

  o Modified the annotation warning, redirecting to the new vignette section
  on how to resolve these.

  o Switched to BiocStyle for the vignette.

CHANGES IN VERSION 1.8.3
-------------------------

BUG FIXES

  o Changed one argument name of the fetchAnnotation function to avoid
  arguments colliding when providing additional arguments to be passed
  to DESeq through the ... argument. The argument has been renamed from
  'method' to 'annotationMethod'.

CHANGES IN VERSION 1.8.2
-------------------------

  o Reworked the parsing of gtf formatted files.

  o Added a few more sanity checks in the easyRNASeq 
  function.

  o Added a use case to the vignette to show how to 
  generate and use synthetic transcripts rather than
  "gene models".

  o Corrected a bug when exporting a SummarizedExperiment
  for "transcrips" and "geneModels" count methods.

CHANGES IN VERSION 1.8.1
-------------------------

	o Imported changes that were lost from version 1.6.4

	o Adapted the dependencies version to match the
	Bioconductor release version 2.13

	o Switched to use GAlignments instead of 
	GappedAlignments

	o Corrected 2 Notes occuring during R CMD check that
	had to do with the DESCRIPTION Depends and Enhances
	fields and the use of pkg:::name vs. pkg::name

	o Added Bastian Schiffthaler as an author

CHANGES IN VERSION 1.8.0
-------------------------

  o No changes, Bioconductor release version 2.13

CHANGES IN VERSION 1.7.0
-------------------------

  o No changes, Bioconductor development version 2.13

CHANGES IN VERSION 1.6.4
-------------------------

  o Made the plotDispLSD function visible (now proporly
exported)

CHANGES IN VERSION 1.6.3
-------------------------

  o Added a plotDispLSD function that adds a density
  estimate color map to the DESeq plotDispEsts function.

CHANGES IN VERSION 1.6.2
-------------------------

  o Released the constraint on the gtf annotation file 
  parsing to allow for gtf file without the "exon_number"
  attribute such as those retrieved from UCSC.

CHANGES IN VERSION 1.6.1
-------------------------

  o Adapted the package dependency

  o Added a check to stop processing whenever 
  annotation are outside of the chromosomes.

CHANGES IN VERSION 1.6.0
-------------------------

  o No changes, Bioconductor release version 2.12

CHANGES IN VERSION 1.5.1
-------------------------
	o Adapted the dependencies version to match the
	Bioconductor release version 2.11

BUG FIXES

    o corrected an innapropriate function call to an internal
    function (as in stable version 1.4.2)

CHANGES IN VERSION 1.5.0
-------------------------

	o No changes, Bioconductor development version 2.12

CHANGES IN VERSION 1.4.2
-------------------------

BUG FIXES

    o corrected an innapropriate function call to an internal function

CHANGES IN VERSION 1.4.1
-------------------------

	o Adapted the dependencies version to match the
	Bioconductor release version 2.11

CHANGES IN VERSION 1.4.0
-------------------------

	o No changes, Bioconductor release version 2.11

CHANGES IN VERSION 1.3.14
-------------------------

NEW FEATURES

    o easyRNASeq now returns a SummarizedExperiment in
    an effort to consolidate the objects used for Next 
    Generation Sequencing in Bioconductor. This is the
    new default of the count function. The count function
    is a new function to supersed easyRNASeq in the coming 
    development version (1.5.x) to consolidate the 
    parameters and output of the easyRNASeq function.

BUG FIXES

    o corrected a validity check that went permissive.
    o changed the print method to display the read length
    range when dealing with variable read lengths rather
    than every single value.

CHANGES IN VERSION 1.3.13
-------------------------

BUG FIXES

    o Same correction as in the stable version 1.2.5, 
    but for those already corrected in version 1.3.3.

CHANGES IN VERSION 1.3.12
-------------------------

    o Providing the 'outputFormat' argument is not
    necessary anymore, it defaults to matrix (i.e. count
    table).

    o Relaxed the gtf file checking. If the gene_name is
    absent, the gene_id is used instead.

    o Improved some reporting and remove a bottle-neck 
    occuring when there are many sequences in the reference.

BUG FIXES

    o Ensure that only the matched ranges are returned when
    reading gapped alignments.

    o The library size is more exactly calculated and is the 
    number of aligned reads.

    o Corrected a bug in the validity checking that prevented
    bam files created by different aligners using the same 
    reference to be processed as the reference sequences were
    not ordered in the same fashion.

CHANGES IN VERSION 1.3.11
-------------------------

BUG FIXES

    o Fixed a bug in the gtf file handling reported
    by Mark Robinson.

CHANGES IN VERSION 1.3.10
-------------------------

    o Some vignette discrepancies have been corrected. 
    Thanks to Richard Friedman for spotting them.

    o Providing the 'filesDirectory' argument is not
    necessary anymore, if the files to proceed are 
    present in the current directory. Indeed, this parameter
    now defaults to the current directory as can be found out
    using 'getwd()'.

BUG FIXES

    o Fixed a bug introduced by a change in the IRanges
    coverage function return value.

CHANGES IN VERSION 1.3.9
-------------------------

    o Added the manuscript citation.

    o Updated the package version dependencies.

BUG FIXES

    o Improved the support for reads of different lengths.

    o A cosmetic change to report read lengths as well when 
    read files with variable read length are processed.

    o Corrected a bug and enhanced the loading of gtf annotation
    files. Thanks to Tomasz Kulinski for spotting the issue and 
    providing the dataset to reproduce it.    

CHANGES IN VERSION 1.3.8
-------------------------
NEW FEATURES

    o Now bam files can be processed in parallel (long time
    request from Wade Davis). If the easyRNASeq argument 'nbCore'
    is greater than 1 (1 being the default), then that many core
    will be used to process the read files in parallel. 
    Pay attention not to use too many cores and have enough 
    memory available. The memory load scales up linearly with
    the number of files processed.

CHANGES IN VERSION 1.3.7
-------------------------
NEW FEATURES

    o easyRNASeq now supports read of different lenghts. Thanks
    to Mark Robinson for the toy dataset.

    o Added a function that lists existing organism conversion
    when applying the validity checks.

    o Added a bp.coverage to the fetchCoverage function that 
    defaults to FALSE. To allow for variable length reads, it
    now returns read coverage proportion per bp by default.

    o Added additional checks in the .checkArguments internal
    function.

BUG FIXES

    o Not a real bug, but more a consolidation. When an organism is
    unknown and no custom.map is provided, then the validity checks
    are turned off and a warning is emitted.

    o Providing the chr.sizes as as list has been deprecated. 
    Only named numeric vector are supported.

    o Removed a now useless warning in the .readGffGtf function.

    o Modified the RPKM function generic to avoid using a 'protected'
    word as argument: i.e. 'unique' was replaced by 'simplify'

CHANGES IN VERSION 1.3.6
-------------------------
NEW FEATURES
    
    o It is now possible to pass arguments to list files through the 
    three dots. I.e. setting recursive=TRUE is now possible.

BUG FIXES

    o Corrected a bug in the .getArguments internal function.

CHANGES IN VERSION 1.3.5
-------------------------

NEW FEATURES

    o bam is now the default format for the easyRNASeq method.
    o chromosome sizes are now extracted from the BAM header 
    when the 'chr.sizes' argument is set to "auto". Thanks to
    Simon Anders for pushing that off my TODO list and the nice
    implementation.

BUG FIXES
    
    o Adapted to an API change of the edgeR package for estimating
    the tagwise dispersion.

CHANGES IN VERSION 1.3.4
-------------------------

NEW FEATURES

    o Added an additional validity check for chromosome names
    Thanks to Simon Anders for generating a reproducible use-case for that.
    Same change as in the stable version 1.2.3

    o Ensure that gtf with non Ensembl ID are correctly parsed as well.


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

    o Converted the package to use Roxygen2, a Doxygen like in-source
    documentation system for generating the RD and NAMESPACE. The 
    original man page were converted using the Rd2roxygen package and
    the resulting in-source documentation manually edited.

NEW FEATURES

    o Added a type accessor for Genome_intervals object
    o Added a coercion to GRangesList from Genome_intervals
    object

BUG FIXES

    o Adapted to the new arguments of the edgeR estimateTagwiseDisp
    function
    o Removed the dispersion.method argument from the plotMeanVar edgeR
    method call as this argument is defunct.

CHANGES IN VERSION 1.3.2
-------------------------

BUG FIXES

    o Corrected a bug that was considering a GTF file as a GFF file.
    Thanks to Simon Anders for spotting this.

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

NEW FEATURES

    o Added an enhanced read length check (same as stable 1.2.1 change)

CHANGES IN VERSION 1.3.0
-------------------------

    o New development version for Bioconductor 2.11

CHANGES IN VERSION 1.2.5
-------------------------

BUG FIXES

    o Corrected a bug in the condition file name checking.
    o When using edgeR, it was not possible to de-activate
    the drawing of the quality assessment plots.
    o Some edgeR changes to the API have been ported to the
    stable R version, should not have occured... The following 
    are changes that adapt to that new API, changes ported from 
    version the easyRNASeq development version 1.3.3...
    o Adapted to the new arguments of the edgeR 
    estimateTagwiseDisp function
    o Removed the dispersion.method argument from the plotMeanVar
    edgeR method call as this argument is defunct.

CHANGES IN VERSION 1.2.4
-------------------------

    o Added the manuscript citation.
    o Updated the package version dependencies.

CHANGES IN VERSION 1.2.3
-------------------------

NEW FEATURES

    o Added an additional validity check for chromosome names
    Thanks to Simon Anders for generating a reproducible use-case for that.

    o Ensure that gtf with non Ensembl ID are correctly parsed as well.

CHANGES IN VERSION 1.2.2
-------------------------

BUG FIXES

    o Corrected a bug that was considering a GTF file as a GFF file.
    Thanks to Simon Anders for spotting this.

CHANGES IN VERSION 1.2.1
-------------------------

NEW FEATURES

    o Added an enhanced read length check

CHANGES IN VERSION 1.2.0
-------------------------

	o Stable version released as part of Bioconductor 2.10

CHANGES IN VERSION 1.1.10
-------------------------

NEW FEATURES

    o Added a naPositionFilter extending ShortRead srFilters

BUG FIXES

    o Worked on Wade Davis case with 3 different sets of chromosome
    names in the three different input (reads, annotation, chromosome
    sizes) 

    o Worked on smoother error handling when not using bam
    files. Again through Wade Davis example 

    o Ensured that chromosome names conversion occurs whether provided
    with a factor or a character vector 
    
    o extended the NAMESPACE

CHANGES IN VERSION 1.1.9
------------------------

BUG FIXES

    o Added an \alias{RNAseq} to ease the class documentation access;
    an H.Pages suggestion 

    o Changed the DESCRIPTION file to make sure that the latest
    ShortRead (1.13.13) and BiocGenerics (0.1.11) package are required 

CHANGES IN VERSION 1.1.8
------------------------

BUG FIXES

    o Corrected the last occurence of the deprecated matchMatrix call

    o Corrected an issue on windows raised by the parallel
    package. Thanks to Wade Davis for pointing that one out. 

CHANGES IN VERSION 1.1.7
------------------------

BUG FIXES

    o Fixed the NAMESPACE and the vignette generation

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

BUG FIXES

    o Thanks to Francesco Lescai, a bug was fixed. Namely I was not
    expecting the chromosome names in the bam files and in the
    chromosome name lists to be two different set with a common
    intersect. I always consider that one would be the subset of the
    other one. Now, when such situation occurs only the common set is
    kept and used for the calculations. 

    o Herve Pages changed the findOverlaps value. It is now an object
    of the Hits class that does not support the matchMatrix accessor
    anymore. The code was adapted to the new accessor queryHits. 

    o Corrected the package structure to add a vignettes
    sub-directory. Moved the relevant files there 

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

BUG FIXES

    o Removed the fitInfo method extension to the DESeq package as it
    as been implemented in that package 

    o Reworked the plotDispersionEstimates and
    .normalizationDispatcher function to deal with the new fitInfo
    function (the information is stored in an environment rather than
    in a list) 

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

    o Package introduced in Bioconductor