January Changes in BioC Devel

    January 2006

adSplit
    New Package:
        This package implements clustering of microarray
        gene expression profiles according to functional
        annotations. For each term genes are annotated to,
        splits into two subclasses are computed and a
        significance of the supporting gene set is determined.       
    Maintainer:
        Claudio Lottaz

affy
    Modification so that main can be specified for image plots
    Fixed typos in man page for ReadAffy()
    Added a cdfname argument to justRMA(), ReadAffy() and read.affybatch() to allow for the use of alternative cdf packages
   
affycoretools
    Updated man page for foldFilt()

affyio
    Updated ability for reading header information
    Fixed some non-fatal memory leaks
    Added additional single cel file parsing ability at C code level
   
affyPLM
    Modification so that main can be specified for image plots
   
AnnBuilder
    Add an extra "\" before "\url{" in SPPkgBuilder() to conform to changes in R-2.3.0
    Removed the makexML part from the ABPrimer vignette. This vignette still needs more work to be compatible with the next release
    Updated yeast related functions
    Added YEASTPFAM, YEASTINTERPRO, YEASTSMART
   
Biobase
    eSet and exprSet validity checking now ensures that sampleNames == row.names of phenoData object
    setReplaceMethod("sampleNames","eSet",...) modified to replace phenoData row names at the same time as sampleNames
    Example dataset 'exSet' renamed sample.exprSet
    Example dataset sample.eSet added

Category
    Added cateGOryMatrix() a helper function for working with GO categories
    Can now also use sparse matrix (graphNEL) representation instead of a matrix to represent category membership - saves lots of memory
    Speeded up cateGOryMatrix()
    Added code and an example for doing MAP category stuff
    Added annotate as a dependency
   
copa
    New Package:
        COPA is a method to find genes that undergo recurrent fusion
             in a given cancer type by finding pairs of genes that have
             mutually exclusive outlier profiles.
    Maintainer:
        James W. MacDonald

EBImage
    Added algorighms to count objects, get object size, etc.   
    All image processing routinges can now directly modify images supplied as arguments without copying them if selected
    Returned 0..1 double data types for grayscales
    Removed argument modify for rotate, scaleImage and sampleImage
   
gcrma
    Added a cdfname to justGCRMA() to allow usage of alternative cdfs
    Removed dependence on reposTools


genefilter
    Bug fixed that occured when 'fac' in rowFtests() contained NA elements
   
geneGraph
    New Package:
        Several utilities to make different plots for genes
    Maintainer:   
        Ting-Yuan Liu

geneplotter
    Subsamples can be used to compute ecdfs with function multiecdf for large data sets to increase graphic performance
    Fixed bug with passing of ... arguments to plot and line
    Fixed signatures of generics and methods
GEOquery
    Removed return()s from parseGEO switch
    Made sure to close connection in getGEO()
    Fixed windows bug by adding mode to download.file() in getGEOfile()
    Fixed error in GDS2eSet where Columns(gds) was called instead of Columns(GDS)
    Fixed typo in parseGSE()
    GDS2eSet now assigns geneNames from the ID_REF column
   
globaltest
    globaltest implemented for multi-valued outcomes
    New function mlogit() to fit multinomial logistic regression
    sort method added for a gt.barplot object
    Replaced old example data with NKI breast cancer data
    multtest() renamed gt.multtest
    Fixed bug in result method for gt.barplot
    Fixed bug that occured when a pathway had all genes NA
    Fixed bug in sort that occured when some pathways had NA p-value
   
goCluster
    Added lookup annotation, fixed GOhyperG comparison in vignette and minor bugfix for FDR class
   
GOstats
    Fixed man page for GOHyperG to account for YEAST
   
graph
    Added doc for nodeData method
    Fixed clusterCoefficient setMethod - no need to use ANY argument
    Improved fromGXL to handle graphs where edgemode is not specified
    Fixed edgeWeights to return a list with the weight vectors named with the appropriate node names instead of node index
    Fixed vignette index so that the attributes vignette is listed properly by openVignette()
    Fixed numEdges for an empty graphNEL (no nodes, no edges)
    Improved fromGXL so attributes are now added to the graph instance
    Attempted to normalize edgeWeights and integrate into generic attribute scheme
    Added attribute serialization to toGXL
    Fixed error message in graphNEL addNode when node is already present
   
graphPart
    New Package:
        The graphPartition package provides tools to partition
                 graphs and hypergraphs.  Currently, interfaces to
                 algorithms in hMETIS, PaToH, and Chaco are provided by
                 this package.
    Maintainer:
        Seth Falcon

iSNetwork
    Added some command line functions, changed how the menus are loaded and changed some of the GUI
    Added man pages
   
limma
    All functions normexp.*() now use simpler arguments:
        'x' instead of 'foreground' and 'background'
        a vector argument 'par' instead of individual parameter arguments
    normexp.fit() now uses a saddlepoint approximation to the normexp density and reverts to the Nelder-Mead optimization method
    New functions dnormexp() and dnormexp.saddle()
    arrayWeights() now returns a vector instead of a matrix
    Bug fix in arrayWeights() - faster and returns more accurate weights
    New function arrayWeightsSimple()
    Subsetting of RGList, MAList and MArrayLM objects by rownames now works correctly even if not all components have rownames
    S3 methods for dimnames<- now defined for RGList and MAList objects
    readTargets() now checks that labels are unique before making row.names
    Files argument of read.maimiges() can now be a data.frame (e.g., the targets frame)
    Bug fix in wtFariables()
    Use of parent = NULL in new.env() removed from makeContrasts() because this is deprecated in R-2.3.0
    topTable() now includes separate columns for raw and adjusted p-values
    When an MArrayLM object is subsetted, the targets frame no longer subsets and the design matrix subsets only on columns
    Previously the design matrix was incorrectly using j to subset rows
    Bug fix to as.data.frame.MArrayLM when row subset is of length 1
    New example using the sma mouse.data for arrayWeights()
    All the linear model functions now use asMatrixWeights(), so that they  will accept vectors of probe or array weights in place of the full matrix of weights
    new function asMatrixWeights() which will convert vectors of probe weights or array weights to a full matrix of weights
    new argument 'maxratio' for arrayWeightsSimple().  This causes the iteration to stop when the ratio of largest to  smallest weights becomes large. The default tol also changed for  arrayweightsSimple().

     
makePlatformDesign
    Bug fix in readSNPCDF()
    Added feature_positon_ID column in PDenvs for NGS
   
metaArray
    Edges in fit.em shrunken to 0.0001

Mfuzz
    New Package:
        Package for noise-robust soft clustering of gene expression data
    Maintainer:
        Matthias Futschik

mmgmos
    Added functions justmmgMOS() and justmgMOS() to use memory more efficiently

MVCClass
    Added a new version of paper 2
   
oligo
    Structural improvements
    PDEnvs are optionally loaded for Affy chips
    Added features to read SD and NPIXELS from CEL files
    Power-user features added in read.celfiles()
    Bug fix rownames in exprs() were in wrong order (CEL files only)
    Bug fix for pm<- mm<-
    Added bg.correct.rma()
   
pdmclass
    Bugfix in pdmGenes()
    Added a NAMESPACE file

RBGL
    More functions return names (node/edge) instead of indices
    Updated vignette
    tsort() output is node names
    More description on HCS parameters in man page for highlyConnSG()
    sp.between returns edge weights for the path, sp.between and dijkstra.sp only take non-negative edge weights
    Fixed ospf GXL so that edge weights are <int> not <string>
    Remade the FileDep graph so that it has up to date class definition from the graph package
   
RdgiPgSQL
    Applied a patch from Mitch Skinner to have dbGetResult() return a more descriptive error
   
RMAPPER
    New Package:
        employs http interface to mapper TFBS database
    Maintainer:
        VJ Carey

RPostgreSQL
    New Package:
        Database interface and PostgreSQL driver for R.
        This version complies with the database interface
        definition as implemented in the package DBI 0.1-4.   
    Maintainer:
        Herve Pages

ScISI
    Added a function to take out protein complexes after QC
    Updated vignette
    Modified getMipsInfo() and constructScISI()
    Removed all the Y2H functions
   
simulatorAPMS
    Removed the following functions and ported them to ScISI:
        compBijection, compareComplex, findSubComp, recCompSize, runAlignment, runCompareComplex

snapCGH
    Added var.fixed arg to fit.model()
    Changed plotSegmentedGenome() so X and Y chromosomes are not plotted by default
   
tilingArray
    Plot method for segmentation now recycles the bpcol argument
    Added more options for the plot method
    Bug fixed in plotSegmentationHeatmap - order of uniq variable corrected
    Moved writeProbeSeqsFasta and seqSequencesWrite to davidTiling