Barolo

• CORR : Added correction suggested by ShenWei-wei to allow support for R v4.0.x
• MOD : Updated README (fixed some links)
• CORR : Updated outdated links to dependences hosted on nextcloud.gustaveroussy.fr

*CloudyMonday*

This version fixes the problem raised by the copynumber::winsorize() function not complying to the new parameter of runmed() to handle NA values.
It also adds some other smaller fixes like fixing a rare desynch of chr names, and some calls to the '%do%' and '%dopar%' operators that were made without loading them from the foreach package.

PreRoscovite2019

• CORR : SNP6.Process(), CSHD.Process() : Edited code to handle changes in the rcnorm package, to discard the "chromosomes" package dependency.
• MOD : Removed some dependencies (already called by other dependencies, like 'copynumber' from 'sequenza') to make installation easier and more convenient.
• MOD : Edited the README.md (rewrote the INSTALL section)
• MOD : WES.Bin() : Added support for BAI files that have the exact same rootname as BAM files (instead of rootname.bam.bai only)
• MOD : Added "call. = FALSE" top all stop() calls
• ADD : WES.Bin() : Added raw read depth plots (to control putative TEST / REF inversion, or sex mismatch)
• MOD : Added content to the LICENCE (default MIT)

Papy60

• Now FACETS can be used on OncoScan, OncoScan_CNV, CytoScan750k and CytoScanHD arrays (still not on SNP6, though).
• Some bugs corrected, too.
• Sequenza CN output fixed.
• Changed the formula for the "most width" ploidy version, so that a value of 0 can't be returned.
• Paving the way to handle non-canonical genomes in the report (not active yet, though).
• Small code changes to adapt to "non-completely covered" genomes (ie, few chromosomes for toy datasets, by example).
• Harmonized the penalty parameter for all segmenters (now just "penalty" rather than "ASCAT.pen", "SEQUENZA.pen" or "FACETS.pen")
• Officially dropping support for sequenza with SNP6 arrays as it leads to a huge RAM consumption by copynumber::aspcf, due to the few probes covering both L2R and BAF.
• Included chromosomes objects in package for hs, mm and rn, to avaoid dependency to another non-public sourceable package.

Trinity

v0.3.3 (20180911) Trinity

• NEW : SEQUENZA segmentation plainly implemented, for both L2R+BAF bivariate segmentation [Segment.SEQUENZA()] AND copy number estimation [ASCN.SEQUENZA()].
• BUG : Segment.ff() : Corrected wrong do.call() call (parameters not given as a list).
• CORR : ASCN.ASCAT() : CN output file was badly formatted.
• MOD : ASCN.ff() : Suppressed the "segmenter" parameter, which is read from the RDS meta$eacon$segmenter value.

v0.3.2-1 (20180906) Bak2Skool

• BUG : WES.Bin() : bin order was totally messed up when capture BED is not properly ordered (desynch due to bin name converted to factor, then levels replaced by mis-ordered rowname). This made WES.Normalize() stop at the GC/wave normalization step due to non-synched bin orders in data when compared to BINpack [Thx to Alexander Lefranc for reporting it] [fixed].
• BUG : ASCN.ff() : fixed a bug in the do.call() call (arguments of ellipsis not converted to a list) [fixed].
• CORR : ASCN.ff() : fixed the handling of out.dir sub() call.
• CORR : README.md : Missing aroma.light dependency [Thx to Alexander Lefranc for reporting it, again].
• ADD : ASCN.ff() : Added a control to check if provided segmentation RDS file is compatible with requested ASCN method.
• MOD : Changed license to MIT.

v0.3.2 (20180808) PapeMamiePichine

• ADD : FACETS segmentation plainly implemented, for both L2R+BAF bivariate segmentation AND copy number estimation. Coincidentaly, some modifications were performed :
• . Segment.ASCAT() and ASCN.ASCAT() wrappers have been renamed to generic Segment() and ASCN(), with a new "segmenter" parameter, to support both of ASCAT and FACETS.
• . Help functions for all segmentation / ASCN functions have been updated.

v0.3.1 (20180731) Recalcitrante

• ADD : New feature : preparing the inclusion of FACETS segmentation methods.
• . This required to include some additional data in the normalized RDS.

v0.3.0-1 (20180725) BzzzzSnap!

• BUG : BINpack.maker() : subfunction failed to load the foreach package [fixed].
• BUG : WES.Bin() : BAM indexation failed [fixed].
• BUG : ASCN.ASCAT.ff() : Typo in dirname replacement made it non functional [fixed].
• BUG : OS.Process() : chromosomes cutting with mingap failed due to calling an inexisting column [fixed].
• ADD : Added a control on list type for functions that require a 'data' object as first parameter.
• DEL : Removed some commented code chunks.

LittleWomanNoCry

• BUG : Segment.SEQUENZA() : added imputation of NA values in L2R object that made copynumber::aspcf() unable to work (happened with microarrays for flagged probes, not WES).
• BUG : Segment.SEQUENZA() : BAF filtering wasn't working properly, resulting in wrong BAF segmentation, for all microarrays.
• BUG : CS.Process.Batch() : wrong variable name in header check.
• CORR : OS.Process() : corrected wrong handling of sex.chr output was forced as c("X", "Y") instead of variable, default c("chrX", "chrY").
• CORR : Segment.FACETS() Segment.SEQUENZA() : Added missing meta 'BAF.filter' in the object.
• CORR : README.md : fixed few links to dependencies, corrected default regex.
• MOD : Segment.* : Changed the structure of the profile PNG filename to "[samplename].SEG.[segmenter].png" (to ease the use of regex for further steps in batch mode).
• MOD : ASCN.ff.Batch() Annotate.ff.Batch() : corrected the default regex.

Bugfixes on v0.3.0

v0.3.0-1 (20180725) BzzzzSnap!

• Bug : BINpack.maker() : subfunction failed to load the foreach package [corrected].
• Bug : WES.Bin() : BAM indexation failed [corrected].
• Bug : ASCN.ASCAT.ff() : Typo in dirname replacement made it non functional [corrected].
• Added a control on list type for functions that require a 'data' object as first parameter.
• Removed some commented code chunks.

PapoQueen

v0.3.0 (20180724) PapoQueen

• WES : Revamped the WES.bin() function to limit RAM consumption and maximize BAM files access when using subthreads.
• WES : Almost completely rewrote the WES.Normalize() function :

• Better L2R normalization using simpler and finer filters.
• Incorporated the segregation of homo / hetero SNPs, using matched normal BAF.
• Added logOR heterozygosity expression for futher compatibility with the FACETS CBS-based bivariate segmenter.

• Discarded the germline prediction from Segment.ASCAT(), now included in each ofthe *.Process() functions (to be more source-specific, mainly for WES).
• All : Removed timestamped dir that made things tricky on some systems (I'm looking at you, Snakemake!)
• All : Removed "EaCoN." prefix from most functions (less self-centric...)
• All : Took care of vectors and columns that could be converted to factor or integer (to free some RAM up).
• All : Added missing support for manual PELT penalty (only asymptotic mode was considered when SER.value was numeric).
• SNP6 : Revamped BAF homozygous calling and rescaling.
• Defined the novel sets of default parameters for all supported technologies.
• Redacted the README.md