[Bioc-devel] NEWS December 2006

James MacDonald jmacdon at med.umich.edu
Tue Jan 9 20:04:30 CET 2007

	December 2006

	Updated writeCdf() to handle the recent name change (block(s)
	to group(s)). Thanks to Ken Simpson
	Added 'legend' argument to plotPCA() to allow end users to not
	add a legend
	Added functionality to pass a matrix to plotPCA()

	Many modifications to convert PLMsets from exprSet to

	Added Illumina annotation package descriptions

	Added makeDataPackage() method for SQLiteAnnDataPkgSeed
	Improved makeDataPackage() - added unlink argument, create
	directory if it doesn't exist
	Added sqliteVersion argument to makeDataPackage(). The default
	is the version of the last available RSQLite package from CRAN
	Added AffySNPPDInfo class
	Code for PDInfo moved to oligo package
	Added db() accessor method

	Improved handling of unknown arguments in eSet's initialize
	method. We put unknown arguments passed in via '...' into
	elements of assayData. We attempt to assign these elements to
	slots and only if no such slot is found does the argument end
	up in assayData. The reason to do this is to allow subclasses
	to rely on default initialize method behavior where extra
	arguments are treated as slot initializers.
	Added more extensive documentation for combine(), AssayData,
	Added methods to annotatedDataFrameFrom(), to initialize from
	'matrix' and NULL
	Allow ExpressionSet to pass '...' through to eSet, where
	arguments are either added as class slots or assigned to
	assayData. Modify eSet handling of '...' so that slot
	arguments are always assigned. Check that assayData members
	are valid as they are assigned to assayData, to avoid cryptic
	error messages.
	Better error message for sampleNames<-, AssayData-method
	Improved error message for createPackage() when destDir
	doesn't exist
	Removed cleanBM() function, fixed bug in getGene() when used
	with mmusculus dataset. Updated getHomolog() so it returns
	both the identifiers used in the query and the identifiers of
	their corresponding homologs (this wil only work from Ensembl
	Enabled getBM() queries without filters
	Added a verbose argument to getBM() to display the XML or
	MySQL queries that are executed
	Fixed formatting of size column in htmlReport output to
	Added validation method to HyperGParams - we now check that
	geneIds and universeGeneIds have no duplicates, that geneIds
	is a subset of universeGeneIds, and that the pvalue cutoff is
	between 0 and 1.
	The progress report window can now be switched off
	More comfortable treatment of plotPlateArgs parameter

	Calls to round() in C code were replaced by NINT() to make
	code more portable
	Problems in the computation of the AIC and BIC criteria were

	New version of EBImage, fully recoded - should now run on
	Windows as well, if the dependencies are satisfied
	The new code is much cleaner and less buggy, using only C
	rather than C++
	Added precompiled windows dll to make compiling on windows
	Added a convenience constructor for rectangular gates called
	Added a %in% method to flowFrame() for rectangular gating
	Changes to filter() to make it a single function that
	dispatches the appropriate %in%
	New vignette that outlines the use of filters
	Added a summary() method primarily for filterResult objects
	Added a print() method for summaryResult objects
	Enabled vector of Entrez Gene mappings as annotation in

	Deprecated getGoGraph - it was meant for internal use only
	Added extensive details to GOstats vignette. Added description
	of input arguments. Showed details of input
	arguments. Explained why parameter object design is
	useful. More details about output and available accessor
	Bug fix to arrayWeights() to correctly ahndle design matrices
	that do not have full column rank
	Updated arrayWeights() man page
	Added new function printtipWeights() that returns sub-array
	quality weights
	New function as.matrix.ExpressionSet()
	Section on array quality weights added to User's guide
	Added PDInfo classes. These represent platform design
	information (chip description with some
	annotation). Subclasses support PDInfo objects backed by
	SQLite DBs, but allow for transitional compatibility with
	environment based PDInfo subclasses.
	Added capablities to use BufferedMatrix objects in assayData
	of *FeatureSet objects (only for Affymetrix now).
	Added read.celfiles2() and rma2() to use BufferedMatrix
	Modified pm() to return a BufferedMatrix.
	Many modifications to get CRLMM working on SQLite-based PDInfo

	Major update. Added llsImpute() to allow better rating and
	comparison of the obtained results, based on Kim et al.,
	Bioinformatics 2005 (for details, see documentation)
	Added nniRes(), a nearest neighbor imputation.
	The threshold in ppca() was changed to 1e-5
	Replaced maxPcs with evalPcs in kEstimate() - user can now
	enter a vector of arbitrary integers
	The pca() function now includes the nni method

	Updated to work with Graphviz 2.10 and 2.12


Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.

More information about the Bioc-devel mailing list