[med-svn] [r-bioc-edger] 01/07: Imported Upstream version 3.2.4~dfsg
Charles Plessy
plessy at alioth.debian.org
Mon Aug 5 10:36:05 UTC 2013
This is an automated email from the git hooks/post-receive script.
plessy pushed a commit to branch master
in repository r-bioc-edger.
commit 543a589e3f2991b4693575dc4d2bbc0eb4ea9d90
Author: Charles Plessy <plessy at debian.org>
Date: Mon Aug 5 19:03:58 2013 +0900
Imported Upstream version 3.2.4~dfsg
---
DESCRIPTION | 6 ++---
NAMESPACE | 13 ++++++-----
R/S3methods.R | 49 +++++++++++++++++++++++++++++++++++++++
R/calcNormFactors.R | 22 ++++++++++++------
R/classes.R | 59 ++++++++++++++---------------------------------
R/cutWithMinN.R | 2 +-
R/dispBinTrend.R | 22 ++++++++++++++----
R/estimateDisp.R | 6 ++++-
R/plotMDS.DGEList.R | 13 +++++------
R/subsetting.R | 42 +++++++++++++++++++++++----------
R/topTags.R | 46 -------------------------------------
inst/NEWS.Rd | 29 +++++++++++++++++++++++
inst/doc/edgeR.pdf | Bin 49354 -> 49354 bytes
man/DGEExact-class.Rd | 32 ++++++++++++++++++--------
man/DGEGLM-class.Rd | 60 ++++++++++++++++++++++++++++++++++++++++++++++++
man/DGEGLM.Rd | 51 ----------------------------------------
man/DGELRT-class.Rd | 57 +++++++++++++++++++++++++++++++++++++++++++++
man/DGELRT.Rd | 46 -------------------------------------
man/DGEList-class.Rd | 11 ++++++---
man/camera.DGEList.Rd | 14 ++++-------
man/cutWithMinN.Rd | 11 +++++----
man/dimnames.Rd | 21 ++++++++---------
man/dispBinTrend.Rd | 3 +--
man/glmfit.Rd | 2 +-
man/plotMDS.DGEList.Rd | 8 +------
man/roast.DGEList.Rd | 29 +++++++----------------
man/topTags.Rd | 19 +++++++++++----
src/core/glm.h | 1 +
28 files changed, 375 insertions(+), 299 deletions(-)
diff --git a/DESCRIPTION b/DESCRIPTION
index 749fda5..dc548c6 100755
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,6 +1,6 @@
Package: edgeR
-Version: 3.2.3
-Date: 2013/04/17
+Version: 3.2.4
+Date: 2013/07/14
Title: Empirical analysis of digital gene expression data in R
Author: Mark Robinson <mrobinson at wehi.edu.au>, Davis McCarthy <dmccarthy at wehi.edu.au>, Yunshun Chen <yuchen at wehi.edu.au>, Aaron Lun <alun at wehi.edu.au>, Gordon Smyth <smyth at wehi.edu.au>
Maintainer: Mark Robinson <mrobinson at wehi.edu.au>, Davis McCarthy
@@ -12,4 +12,4 @@ biocViews: Bioinformatics, DifferentialExpression, SAGE,
HighThroughputSequencing, RNAseq, ChIPseq
Description: Differential expression analysis of RNA-seq and digital gene expression profiles with biological replication. Uses empirical Bayes estimation and exact tests based on the negative binomial distribution. Also useful for differential signal analysis with other types of genome-scale count data.
License: GPL (>=2)
-Packaged: 2013-04-18 05:25:33 UTC; biocbuild
+Packaged: 2013-07-15 04:54:44 UTC; biocbuild
diff --git a/NAMESPACE b/NAMESPACE
index d2ecc8d..c3de40b 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -4,7 +4,7 @@ useDynLib(edgeR)
# All functions exported other than those starting with "."
exportPattern("^[^\\.]")
-exportClasses("DGEList","DGEExact","TopTags","DGEGLM","DGELRT")
+exportClasses("DGEList","DGEExact","DGEGLM","DGELRT","TopTags")
exportMethods("show")
import(limma)
@@ -12,15 +12,16 @@ import(limma)
S3method(as.matrix,DGEList)
S3method(dim,DGEList)
S3method(dim,DGEExact)
-S3method(dim,TopTags)
S3method(dim,DGEGLM)
S3method(dim,DGELRT)
+S3method(dim,TopTags)
S3method(dimnames,DGEList)
-S3method("dimnames<-",DGEList)
+S3method(dimnames,DGEExact)
S3method(dimnames,DGEGLM)
-S3method("dimnames<-",DGEGLM)
+S3method(dimnames,DGELRT)
S3method(dimnames,TopTags)
-S3method("dimnames<-",TopTags)
+S3method("dimnames<-",DGEList)
+S3method("dimnames<-",DGEGLM)
S3method(length,DGEList)
S3method(length,DGEExact)
S3method(length,TopTags)
@@ -29,5 +30,5 @@ S3method(length,DGELRT)
S3method(plotMDS,DGEList)
S3method(as.data.frame,TopTags)
S3method(roast,DGEList)
-S3method(camera,DGEList)
S3method(mroast,DGEList)
+S3method(camera,DGEList)
diff --git a/R/S3methods.R b/R/S3methods.R
new file mode 100644
index 0000000..45df26e
--- /dev/null
+++ b/R/S3methods.R
@@ -0,0 +1,49 @@
+# S3 as.matrix method
+
+as.matrix.DGEList <- function(x,...) as.matrix(x$counts)
+
+# S3 as.data.frame method
+
+as.data.frame.TopTags <- function(x,row.names=NULL,optional=FALSE,...)
+{
+ if(!is.null(row.names)) row.names(x$table) <- row.names
+ x$table
+}
+
+# S3 dim methods
+# These enable nrow() and ncol() as well
+
+dim.DGEList <- function(x) if(is.null(x$counts)) c(0,0) else dim(as.matrix(x$counts))
+dim.DGEGLM <- function(x) if(is.null(x$coefficients)) c(0,0) else dim(as.matrix(x$coefficients))
+dim.DGEExact <- dim.TopTags <- dim.DGELRT <- function(x) if(is.null(x$table)) c(0,0) else dim(as.matrix(x$table))
+
+# S3 length methods
+
+length.DGEList <- length.DGEExact <- length.TopTags <- length.DGEGLM <- length.DGELRT <- function(x) prod(dim(x))
+
+# S3 dimnames methods
+# These enable rownames() and colnames() as well
+
+dimnames.DGEList <- function(x) dimnames(x$counts)
+dimnames.DGEGLM <- function(x) dimnames(x$coefficients)
+dimnames.DGEExact <- dimnames.DGELRT <- dimnames.TopTags <- function(x) dimnames(x$table)
+
+# S3 dimnames<- methods
+# These enable rownames()<- and colnames()<- as well
+
+assign("dimnames<-.DGEList",function(x,value)
+{
+ dimnames(x$counts) <- value
+ if(!is.null(x$samples)) row.names(x$samples) <- value[[2]]
+ if(!is.null(x$genes)) row.names(x$genes) <- value[[1]]
+ x
+})
+
+assign("dimnames<-.DGEGLM",function(x,value)
+{
+ dimnames(x$coefficients) <- value
+ if(!is.null(x$samples)) row.names(x$samples) <- value[[2]]
+ if(!is.null(x$genes)) row.names(x$genes) <- value[[1]]
+ x
+})
+
diff --git a/R/calcNormFactors.R b/R/calcNormFactors.R
index 223762d..2c90dd4 100644
--- a/R/calcNormFactors.R
+++ b/R/calcNormFactors.R
@@ -1,7 +1,7 @@
calcNormFactors <- function(object, method=c("TMM","RLE","upperquartile","none"), refColumn=NULL, logratioTrim=.3, sumTrim=0.05, doWeighting=TRUE, Acutoff=-1e10, p=0.75)
# Scale normalization of RNA-Seq data.
# Mark Robinson. Edits by Gordon Smyth.
-# Created October 22 October 2009. Last modified 24 Feb 2013.
+# Created October 22 October 2009. Last modified 16 Apr 2013.
{
# Check object
if(is(object,"DGEList")) {
@@ -26,9 +26,16 @@ calcNormFactors <- function(object, method=c("TMM","RLE","upperquartile","none")
f <- switch(method,
TMM = {
f75 <- .calcFactorQuantile(data=x, lib.size=lib.size, p=0.75)
- refColumn <- which.min(abs(f75-mean(f75)))
- if(length(refColumn)==0) refColumn <- 1
- apply(x,2,.calcFactorWeighted,ref=x[,refColumn], logratioTrim=logratioTrim, sumTrim=sumTrim, doWeighting=doWeighting, Acutoff=Acutoff)
+ if( is.null(refColumn) )
+ refColumn <- which.min(abs(f75-mean(f75)))
+ if(length(refColumn)==0 | refColumn < 1 | refColumn > ncol(x)) refColumn <- 1
+ f <- rep(NA,ncol(x))
+ for(i in 1:ncol(x))
+ f[i] <- .calcFactorWeighted(obs=x[,i],ref=x[,refColumn], libsize.obs=lib.size[i],
+ libsize.ref=lib.size[refColumn],
+ logratioTrim=logratioTrim, sumTrim=sumTrim,
+ doWeighting=doWeighting, Acutoff=Acutoff)
+ f
},
RLE = .calcFactorRLE(x)/lib.size,
upperquartile = .calcFactorQuantile(x,lib.size,p=p),
@@ -63,15 +70,16 @@ calcNormFactors <- function(object, method=c("TMM","RLE","upperquartile","none")
# f/exp(mean(log(f)))
}
-.calcFactorWeighted <- function(obs, ref, logratioTrim=.3, sumTrim=0.05, doWeighting=TRUE, Acutoff=-1e10)
+.calcFactorWeighted <- function(obs, ref, libsize.obs=NULL, libsize.ref=NULL, logratioTrim=.3, sumTrim=0.05, doWeighting=TRUE, Acutoff=-1e10)
{
if( all(obs==ref) ) return(1)
obs <- as.numeric(obs)
ref <- as.numeric(ref)
- nO <- sum(obs)
- nR <- sum(ref)
+ if( is.null(libsize.obs) ) nO <- sum(obs) else nO <- libsize.obs
+ if( is.null(libsize.ref) ) nR <- sum(ref) else nR <- libsize.ref
+
logR <- log2((obs/nO)/(ref/nR)) # log ratio of expression, accounting for library size
absE <- (log2(obs/nO) + log2(ref/nR))/2 # absolute expression
v <- (nO-obs)/nO/obs + (nR-ref)/nR/ref # estimated asymptotic variance
diff --git a/R/classes.R b/R/classes.R
index 837a4dd..969638c 100644
--- a/R/classes.R
+++ b/R/classes.R
@@ -1,12 +1,12 @@
require(methods)
+# S4 classes
+
setClass("DGEExact",
-# Linear model fit
representation("list")
)
setClass("DGEList",
-# Linear model fit
representation("list")
)
@@ -18,50 +18,25 @@ setClass("DGELRT",
representation("list")
)
+setClass("TopTags",
+representation("list")
+)
+
+# Set inheritance
+# The LargeDataObject class is set in limma and provides a show method
+
setIs("DGEList","LargeDataObject")
setIs("DGEExact","LargeDataObject")
setIs("DGEGLM","LargeDataObject")
setIs("DGELRT","LargeDataObject")
-dim.DGEList <- function(x) if(is.null(x$counts)) c(0, 0) else dim(as.matrix(x$counts))
-dim.DGEGLM <- function(x) if(is.null(x$coefficients)) c(0, 0) else dim(as.matrix(x$coefficients))
-dim.DGEExact <- dim.TopTags <- dim.DGELRT <- function(x) if(is.null(x$table)) c(0, 0) else dim(as.matrix(x$table))
-
-length.DGEList <- length.DGEExact <- length.TopTags <- length.DGEGLM <- length.DGELRT <- function(x) prod(dim(x))
-
-dimnames.DGEList <- function(x) dimnames(x$counts)
-assign("dimnames<-.DGEList",function(x,value)
-{
- dimnames(x$counts) <- value
- if(!is.null(x$samples)) row.names(x$samples) <- value[[2]]
- if(!is.null(x$genes)) row.names(x$genes) <- value[[1]]
- x
-})
-
-dimnames.DGEExact <- function(x) dimnames(x$coefficients)
-assign("dimnames<-.DGEExact",function(x,value)
-{
- dimnames(x$coefficients) <- value
- if(!is.null(x$samples)) row.names(x$samples) <- value[[2]]
- if(!is.null(x$genes)) row.names(x$genes) <- value[[1]]
- x
-})
+# Show method
-dimnames.DGEGLM <- function(x) dimnames(x$coefficients)
-assign("dimnames<-.DGEGLM",function(x,value)
-{
- dimnames(x$coefficients) <- value
- if(!is.null(x$samples)) row.names(x$samples) <- value[[2]]
- if(!is.null(x$genes)) row.names(x$genes) <- value[[1]]
- x
+setMethod("show", "TopTags", function(object) {
+ if(object$test=="exact") {
+ cat("Comparison of groups: ",paste(rev(object$comparison),collapse="-"),"\n")
+ } else {
+ cat("Coefficient: ",object$comparison,"\n")
+ }
+ print(object$table)
})
-
-dimnames.TopTags <- function(x) dimnames(x$table)
-assign("dimnames<-.TopTags",function(x,value)
-{
- dimnames(x$table) <- value
- x
-})
-
-as.matrix.DGEList <- function(x,...) as.matrix(x$counts)
-
diff --git a/R/cutWithMinN.R b/R/cutWithMinN.R
index 4dfd273..c3ecf3d 100644
--- a/R/cutWithMinN.R
+++ b/R/cutWithMinN.R
@@ -58,7 +58,7 @@ cutWithMinN <- function(x, intervals=2, min.n=1)
n <- floor(nx/intervals)
nresid <- nx - intervals*n
n <- rep.int(n,intervals)
- n[1] <- n[1] + nresid
+ if(nresid>0) n[1:nresid] <- n[1:nresid]+1
z <- rep(1:intervals,n)
z[o] <- z
return(list(group=z,breaks=breaks.eqn))
diff --git a/R/dispBinTrend.R b/R/dispBinTrend.R
index 8c72f64..7406a4d 100644
--- a/R/dispBinTrend.R
+++ b/R/dispBinTrend.R
@@ -1,8 +1,8 @@
-dispBinTrend <- function(y, design=NULL, offset=NULL, df=5, span=0.3, min.n=400, method.bin="CoxReid", method.trend="spline", trace=0, AveLogCPM=NULL, ...)
+dispBinTrend <- function(y, design=NULL, offset=NULL, df=5, span=0.3, min.n=400, method.bin="CoxReid", method.trend="spline", AveLogCPM=NULL, ...)
# Estimate common dispersion in bins based on AveLogCPM,
# then fit a curve through the dispersions
# Davis McCarthy, Gordon Smyth
-# Created 10 Feb 2011. Last modified 11 March 2013.
+# Created 10 Feb 2011. Last modified 17 April 2013.
{
# Check y
y <- as.matrix(y)
@@ -32,8 +32,22 @@ dispBinTrend <- function(y, design=NULL, offset=NULL, df=5, span=0.3, min.n=400,
# Define bins of genes based on min.n in each bin
# All zero rows are marked as group==0
group <- as.numeric(pos)
- nbins <- floor(npostags/min.n)
- nbins <- min(max(nbins,1),1000)
+
+ if(npostags < 100)
+ nbins <- 1
+ else {
+ nbins <- floor(npostags^0.4)
+ nbins <- min(nbins,1000)
+ min.n <- min(min.n,floor(npostags/nbins))
+ }
+ if(min.n < 50) {
+ nbins <- floor(npostags/50)
+ min.n <- 50
+ }
+
+# nbins <- floor(npostags/min.n)
+# nbins <- min(max(nbins,1),1000)
+
if(nbins>1) group[pos] <- cutWithMinN(AveLogCPM[pos],intervals=nbins,min.n=min.n)$group
# Estimate dispersion in each bin
diff --git a/R/estimateDisp.R b/R/estimateDisp.R
index e6446ad..9509e2d 100644
--- a/R/estimateDisp.R
+++ b/R/estimateDisp.R
@@ -28,7 +28,11 @@ estimateDisp <- function(y, design=NULL, offset=NULL, prior.df=NULL, trend.metho
# Classic edgeR
if(is.null(design)){
- design <- model.matrix(~group)
+ # One group
+ if(length(levels(group))==1)
+ design <- matrix(1,nlibs,1)
+ else
+ design <- model.matrix(~group)
if( all(tabulate(group)<=1) ) {
warning("There is no replication, setting dispersion to NA.")
y$common.dispersion <- NA
diff --git a/R/plotMDS.DGEList.R b/R/plotMDS.DGEList.R
index 7c007e1..6c9b4c5 100644
--- a/R/plotMDS.DGEList.R
+++ b/R/plotMDS.DGEList.R
@@ -1,16 +1,15 @@
plotMDS.DGEList <- function (x, top=500, labels=colnames(x), col=NULL, cex=1, dim.plot=c(1, 2), ndim=max(dim.plot), xlab=NULL, ylab=NULL, method="logFC", prior.count=2, gene.selection="pairwise", ...)
# Multidimensional scaling plot of digital gene expression profiles
# Yunshun Chen, Mark Robinson and Gordon Smyth
-# 23 May 2011. Last modified 7 Feb 2013.
+# 23 May 2011. Last modified 28 May 2013.
{
-# Remove rows with missing or Inf values
- ok <- is.finite(x$counts)
- if(!all(ok)) {
- x <- x[rowSums(ok)>0,]
- x$samples$lib.size <- rowSums(x$counts)
- }
+# Check x
+ x$counts <- as.matrix(x$counts)
+ if(!all(is.finite(x$counts))) stop("Missing or infinite counts not allowed")
+
nprobes <- nrow(x)
nsamples <- ncol(x)
+ if(nsamples < 3) stop("Need at least 3 columns of data")
# Check value for labels
if(is.null(labels)) labels <- 1:nsamples
diff --git a/R/subsetting.R b/R/subsetting.R
index 7551b0a..8945612 100644
--- a/R/subsetting.R
+++ b/R/subsetting.R
@@ -54,9 +54,10 @@ function(object, i, j, ...) {
assign("[.DGEGLM",
function(object, i, j, ...)
# Subsetting for DGEGLM objects
-# Davis McCarthy
-# 11 May 2011. Last modified 5 Dec 2012.
+# Davis McCarthy, Gordon Smyth
+# 11 May 2011. Last modified 8 April 2013.
{
+ if(nargs() != 3) stop("Two subscripts required",call.=FALSE)
if(!missing(j))
stop("Subsetting columns not allowed for DGEGLM object. Try subsetting elements of DGEGLM object instead.",call.=FALSE)
if(!missing(i)) {
@@ -79,9 +80,10 @@ function(object, i, j, ...)
assign("[.DGEExact",
function(object, i, j, ...)
# Subsetting for DGEExact objects
-# Davis McCarthy
-# 6 October 2010. Last modified 6 Oct 2010.
+# Davis McCarthy, Gordon Smyth
+# 6 October 2010. Last modified 8 April 2013.
{
+ if(nargs() != 3) stop("Two subscripts required",call.=FALSE)
if(!missing(j))
stop("Subsetting columns not allowed for DGEExact object. Try subsetting object$table instead.",call.=FALSE)
if(!missing(i)) {
@@ -91,12 +93,14 @@ function(object, i, j, ...)
object
})
+
assign("[.DGELRT",
function(object, i, j, ...)
# Subsetting for DGELRT objects
-# Davis McCarthy
-# 6 April 2011. Last modified 23 June 2011.
+# Davis McCarthy, Gordon Smyth
+# 6 April 2011. Last modified 8 April 2013.
{
+ if(nargs() != 3) stop("Two subscripts required",call.=FALSE)
if(!missing(j))
stop("Subsetting columns not allowed for DGELRT object. Try subsetting object$table instead.",call.=FALSE)
if(!missing(i)) {
@@ -112,15 +116,29 @@ function(object, i, j, ...)
})
-
assign("[.TopTags",
function(object, i, j, ...)
# Subsetting for TopTags objects
-# Gordon Smyth
-# 7 October 2009. Last modified 9 October 2009.
+# Gordon Smyth
+# 7 October 2009. Last modified 8 April 2013.
{
- if(!missing(j)) stop("Subsetting columns not allowed for TopTags object. Try subsetting object$table instead.",call.=FALSE)
- if(!missing(i)) object$table <- object$table[i,,drop=FALSE]
+ if(nargs() != 3) stop("Two subscripts required",call.=FALSE)
+ if(missing(i))
+ if(missing(j))
+ return(object)
+ else {
+ object$table <- object$table[,j,drop=FALSE]
+ }
+ else {
+ if(is.character(i)) {
+ i <- match(i,rownames(object$counts))
+ i <- i[!is.na(i)]
+ }
+ if(missing(j)) {
+ object$table <- object$table[i,,drop=FALSE]
+ } else {
+ object$table <- object$table[i,j,drop=FALSE]
+ }
+ }
object
})
-
diff --git a/R/topTags.R b/R/topTags.R
index f69b3a4..bf8241c 100644
--- a/R/topTags.R
+++ b/R/topTags.R
@@ -1,49 +1,3 @@
-setClass("TopTags",
-representation("list")
-)
-
-setMethod("show", "TopTags", function(object) {
- if(object$test=="exact") {
- cat("Comparison of groups: ",paste(rev(object$comparison),collapse="-"),"\n")
- } else {
- cat("Coefficient: ",object$comparison,"\n")
- }
- print(object$table)
-})
-
-assign("[.TopTags",
-function(object, i, j, ...) {
-# Subsetting for TopTags objects
-# Gordon Smyth
-# 9 Feb 2012. Last modified 9 Feb 2012.
-
- if(nargs() != 3) stop("Two subscripts required",call.=FALSE)
- if(missing(i))
- if(missing(j))
- return(object)
- else {
- object$table <- object$table[,j,drop=FALSE]
- }
- else {
- if(is.character(i)) {
- i <- match(i,rownames(object$counts))
- i <- i[!is.na(i)]
- }
- if(missing(j)) {
- object$table <- object$table[i,,drop=FALSE]
- } else {
- object$table <- object$table[i,j,drop=FALSE]
- }
- }
- object
-})
-
-as.data.frame.TopTags <- function(x,row.names=NULL,optional=FALSE,...)
-{
- if(!is.null(row.names)) row.names(x$table) <- row.names
- x$table
-}
-
topTags <- function(object,n=10,adjust.method="BH",sort.by="PValue")
# Summary table of the n most differentially expressed tags
# Mark Robinson, Davis McCarthy, Gordon Smyth
diff --git a/inst/NEWS.Rd b/inst/NEWS.Rd
index 8bd8672..244825c 100644
--- a/inst/NEWS.Rd
+++ b/inst/NEWS.Rd
@@ -2,6 +2,32 @@
\title{edgeR News}
\encoding{UTF-8}
+\section{Version 3.2.4}{\itemize{
+\item
+Refinement to cutWithMinN() to make the bin numbers more equal in the worst case.
+
+\item
+estimateDisp() now creates the design matrix correctly when the design matrix is not given as an argument and there is only one group. Previously this case gave an error.
+
+\item
+plotMDS.DGEList now gives a friendly error message when there are fewer than 3 data columns.
+
+\item
+Refinement to computation for nbins in dispBinTrend. Now changes more smoothly with the number of genes. trace argument is retired.
+
+\item
+Fixes to calcNormFactors with method="TMM" so that it takes account of lib.size and refCol if these are preset.
+
+\item
+Some fixes and cleaning up of subsetting code.
+
+\item
+Minor edit to glm.h code.
+
+\item
+Edits to camera.DGEList, cutWithMinN and plotMDS.DGEList help pages. Edits to help pages for the data classes.
+}}
+
\section{Version 3.2.3}{\itemize{
\item
Update to cutWithMinN() so that it does not fail even when there are many repeated x values.
@@ -30,6 +56,9 @@ The function estimateDisp() provides a simpler alternative pipeline and in princ
It can also incorporate automatic estimation of the prior degrees of freedom, and can do this in a robust fashion.
\item
+Default prior.df for estimateTagwiseDisp() and estimateGLMTagwiseDisp() reduced from 20 to 10.
+
+\item
glmLRT() now permits the contrast argument to be a matrix with multiple columns, making the treatment of this argument analogous to that of the coef argument.
\item
diff --git a/inst/doc/edgeR.pdf b/inst/doc/edgeR.pdf
index 594b88e..f4c429d 100644
Binary files a/inst/doc/edgeR.pdf and b/inst/doc/edgeR.pdf differ
diff --git a/man/DGEExact-class.Rd b/man/DGEExact-class.Rd
index cdc76e8..0232040 100644
--- a/man/DGEExact-class.Rd
+++ b/man/DGEExact-class.Rd
@@ -5,25 +5,39 @@
\title{differential expression of Digital Gene Expression data - class}
\description{
-A simple list-based class for storing results of differential expression analysis for DGE data
+A list-based S4 class for for storing results of a differential expression analysis for DGE data.
}
-\section{Slots/List Components}{
-Objects of this class contain the following list components:
+\section{List Components}{
+For objects of this class, rows correspond to genomic features and columns to statistics associated with the differential expression analysis.
+The genomic features are called genes, but in reality might correspond to transcripts, tags, exons etc.
- \code{table}: data frame containing the log-concentration (i.e. expression level), the log-fold change in expression between the two groups/conditions and the exact p-value for differential expression, for each tag.
+Objects of this class contain the following list components:
+\tabular{ll}{
+ \code{table } \tab data frame containing columns for the log2-fold-change, \code{logFC}, the average log2-counts-per-million, \code{logCPM}, and the two-sided p-value \code{PValue}.\cr
- \code{comparison}: vector giving the two experimental groups/conditions being compared.
+ \code{comparison } \tab vector giving the two experimental groups/conditions being compared.\cr
- \code{genes}: a data frame containing information about each transcript (can be \code{NULL}).
+ \code{genes } \tab a data frame containing information about each gene (can be \code{NULL}).\cr
+}
}
\section{Methods}{
-This class inherits directly from class \code{list} so any operation appropriate for lists will work on objects of this class.
-\code{DGEExact} objects also have a \code{show} method.
+This class inherits directly from class \code{list}, so \code{DGEExact} objects can be manipulated as if they were ordinary lists.
+However they can also be treated as if they were matrices for the purposes of subsetting.
+
+The dimensions, row names and column names of a \code{DGEExact} object are defined by those of \code{table}, see \code{\link{dim.DGEExact}} or \code{\link{dimnames.DGEExact}}.
+
+\code{DGEExact} objects can be subsetted, see \code{\link{subsetting}}.
+
+\code{DGEExact} objects also have a \code{show} method so that printing produces a compact summary of their contents.
}
-\author{Mark Robinson, Davis McCarthy}
+\author{edgeR team. First created by Mark Robinson and Davis McCarthy}
+
+\seealso{
+Other classes defined in edgeR are \code{\link{DGEList-class}}, \code{\link{DGEGLM-class}}, \code{\link{DGELRT-class}}, \code{\link{TopTags-class}}
+}
\keyword{classes}
diff --git a/man/DGEGLM-class.Rd b/man/DGEGLM-class.Rd
new file mode 100644
index 0000000..73f8071
--- /dev/null
+++ b/man/DGEGLM-class.Rd
@@ -0,0 +1,60 @@
+\name{DGEGLM-class}
+\docType{class}
+\alias{DGEGLM-class}
+\alias{show,DGEGLM-method}
+\title{Digital Gene Expression Generalized Linear Model results - class}
+
+\description{
+A list-based S4 class for storing results of a GLM fit to each gene in a DGE dataset.
+}
+
+\section{List Components}{
+For objects of this class, rows correspond to genomic features and columns to coefficients in the linear model.
+The genomic features are called genes, but in reality might correspond to transcripts, tags, exons etc.
+
+Objects of this class contain the following list components:
+\tabular{ll}{
+ \code{coefficients } \tab matrix containing the coefficients computed from fitting the model defined by the design matrix to each gene in the dataset.\cr
+
+ \code{df.residual } \tab vector containing the residual degrees of freedom for the model fit to each gene in the dataset.\cr
+
+ \code{deviance } \tab vector giving the deviance from the model fit to each gene.\cr
+
+ \code{design } \tab design matrix for the full model from the likelihood ratio test.\cr
+
+ \code{offset } \tab scalar, vector or matrix of offset values to be included in the GLMs for each gene.\cr
+
+ \code{samples } \tab data frame containing information about the samples comprising the dataset.\cr
+
+ \code{genes } \tab data frame containing information about the genes or tags for which we have DGE data (can be \code{NULL} if there is no information available).\cr
+
+ \code{dispersion } \tab scalar or vector providing the value of the dispersion parameter used in the negative binomial GLM for each gene.\cr
+
+ \code{lib.size } \tab vector providing the effective library size for each sample in the dataset.\cr
+
+ \code{weights } \tab matrix of weights used in the GLM fitting for each gene.\cr
+
+ \code{fitted.values } \tab the fitted (expected) values from the GLM for each gene.\cr
+
+ \code{AveLogCPM } \tab numeric vector giving average log2 counts per million for each gene.
+}
+}
+
+\section{Methods}{
+This class inherits directly from class \code{list} so any operation appropriate for lists will work on objects of this class.
+
+The dimensions, row names and column names of a \code{DGEGLM} object are defined by those of the dataset, see \code{\link{dim.DGEGLM}} or \code{\link{dimnames.DGEGLM}}.
+
+\code{DGEGLM} objects can be subsetted, see \code{\link{subsetting}}.
+
+\code{DGEGLM} objects also have a \code{show} method so that printing produces a compact summary of their contents.
+
+}
+\author{edgeR team. First created by Davis McCarthy.}
+
+\seealso{
+Other classes defined in edgeR are \code{\link{DGEList-class}}, \code{\link{DGEExact-class}}, \code{\link{DGELRT-class}}, \code{\link{TopTags-class}}
+}
+
+\keyword{classes}
+
diff --git a/man/DGEGLM.Rd b/man/DGEGLM.Rd
deleted file mode 100644
index 6ba4d3b..0000000
--- a/man/DGEGLM.Rd
+++ /dev/null
@@ -1,51 +0,0 @@
-\name{DGEGLM-class}
-\docType{class}
-\alias{DGEGLM-class}
-\alias{show,DGEGLM-method}
-\title{Digital Gene Expression Generalized Linear Model results - class}
-
-\description{
-A simple list-based class for storing results of a GLM fit
-to each tag/gene in a DGE dataset.
-}
-
-\section{Slots/List Components}{
-Objects of this class contain the following list components:
-
- \code{coefficients}: matrix containing the coefficients
- computed from fitting the model defined by the design matrix to each gene/tag in the dataset.
-
- \code{df.residual}: vector containing the residual degrees of freedom for the model fit to each tag/gene in the dataset.
-
- \code{deviance}: vector giving the deviance from the model fit to each tag/gene.
-
- \code{design}: design matrix for the full model from the likelihood
- ratio test.
-
- \code{offset}: scalar, vector or matrix of offset values to be included in the GLMs for each tag/gene.
-
- \code{samples}: data frame containing information about the samples comprising the dataset.
-
- \code{genes}: data frame containing information about the genes or tags for which we have DGE data (can be \code{NULL} if there is no information available).
-
- \code{dispersion}: scalar or vector providing the value of the dispersion parameter used in the negative binomial GLM for each tag/gene.
-
- \code{lib.size}: vector providing the effective library size for each sample in the dataset.
-
- \code{weights}: matrix of weights used in the GLM fitting for each tag/gene.
-
- \code{fitted.values}: the fitted (expected) values--here they are counts--from the GLM for each tag/gene.
-
- \code{abundance}: vector of gene/tag abundances (expression level), on the log2 scale, computed from the mean count for each gene/tag after scaling count by normalized library size.
-
-}
-
-\section{Methods}{
-This class inherits directly from class \code{list} so any operation appropriate for lists will work on objects of this class.
-\code{DGEGLM} objects also have a \code{show} method.
-}
-
-\author{Davis McCarthy}
-
-\keyword{classes}
-
diff --git a/man/DGELRT-class.Rd b/man/DGELRT-class.Rd
new file mode 100644
index 0000000..792eabe
--- /dev/null
+++ b/man/DGELRT-class.Rd
@@ -0,0 +1,57 @@
+\name{DGELRT-class}
+\docType{class}
+\alias{DGELRT-class}
+\alias{show,DGELRT-method}
+\title{Digital Gene Expression Likelihood Ratio Test data and results - class}
+
+\description{
+A list-based S4 class for storing results of a GLM-based differential expression analysis for DGE data.
+}
+
+\section{List Components}{
+For objects of this class, rows correspond to genomic features and columns to statistics associated with the differential expression analysis.
+The genomic features are called genes, but in reality might correspond to transcripts, tags, exons etc.
+
+Objects of this class contain the following list components:
+\tabular{ll}{
+ \code{table } \tab data frame containing the log-concentration (i.e. expression level), the log-fold change in expression between the two groups/conditions and the exact p-value for differential expression, for each gene.\cr
+
+ \code{coefficients.full } \tab matrix containing the coefficients
+ computed from fitting the full model (fit using \code{glmFit} and a
+ given design matrix) to each gene in the dataset.\cr
+
+ \code{coefficients.null } \tab matrix containing the coefficients
+ computed from fitting the null model to each gene in the
+ dataset. The null model is the model to which the full model is
+ compared, and is fit using \code{glmFit} and dropping selected
+ column(s) (i.e. coefficient(s)) from the design matrix for the full model.\cr
+
+ \code{design } \tab design matrix for the full model from the likelihood
+ ratio test.\cr
+
+ \code{... } \tab if the argument \code{y} to \code{glmLRT} (which
+ produces the \code{DGELRT} object) was itself a \code{DGEList} object, then
+ the \code{DGELRT} will contain all of the elements of \code{y},
+ except for the table of counts and the table of pseudocounts.\cr
+}
+}
+
+\section{Methods}{
+This class inherits directly from class \code{list}, so \code{DGELRT} objects can be manipulated as if they were ordinary lists.
+However they can also be treated as if they were matrices for the purposes of subsetting.
+
+The dimensions, row names and column names of a \code{DGELRT} object are defined by those of \code{table}, see \code{\link{dim.DGELRT}} or \code{\link{dimnames.DGELRT}}.
+
+\code{DGELRT} objects can be subsetted, see \code{\link{subsetting}}.
+
+\code{DGELRT} objects also have a \code{show} method so that printing produces a compact summary of their contents.
+}
+
+\author{edgeR team. First created by Davis McCarthy}
+
+\seealso{
+Other classes defined in edgeR are \code{\link{DGEList-class}}, \code{\link{DGEExact-class}}, \code{\link{DGEGLM-class}}, \code{\link{TopTags-class}}
+}
+
+\keyword{classes}
+
diff --git a/man/DGELRT.Rd b/man/DGELRT.Rd
deleted file mode 100644
index 15dd35c..0000000
--- a/man/DGELRT.Rd
+++ /dev/null
@@ -1,46 +0,0 @@
-\name{DGELRT-class}
-\docType{class}
-\alias{DGELRT-class}
-\alias{show,DGELRT-method}
-\title{Digital Gene Expression Likelihood Ratio Test data and results - class}
-
-\description{
-A simple list-based class for storing results of a GLM-based
-differential expression analysis for DGE data, with evidence for
-differential expression assessed using a likelihood ratio test.
-}
-
-\section{Slots/List Components}{
-Objects of this class contain the following list components:
-
- \code{table}: data frame containing the log-concentration (i.e. expression level), the log-fold change in expression between the two groups/conditions and the exact p-value for differential expression, for each tag.
-
- \code{coefficients.full}: matrix containing the coefficients
- computed from fitting the full model (fit using \code{glmFit} and a
- given design matrix) to each gene/tag in the dataset.
-
- \code{coefficients.null}: matrix containing the coefficients
- computed from fitting the null model to each gene/tag in the
- dataset. The null model is the model to which the full model is
- compared, and is fit using \code{glmFit} and dropping selected
- column(s) (i.e. coefficient(s)) from the design matrix for the full model.
-
- \code{design}: design matrix for the full model from the likelihood
- ratio test.
-
- \code{...}: if the argument \code{y} to \code{glmLRT} (which
- produces the \code{DGELRT} object) was itself a \code{DGEList} object, then
- the \code{DGELRT} will contain all of the elements of \code{y},
- except for the table of counts and the table of pseudocounts.
-
-}
-
-\section{Methods}{
-This class inherits directly from class \code{list} so any operation appropriate for lists will work on objects of this class.
-\code{DGELRT} objects also have a \code{show} method.
-}
-
-\author{Davis McCarthy}
-
-\keyword{classes}
-
diff --git a/man/DGEList-class.Rd b/man/DGEList-class.Rd
index d4dad5f..d3afb0f 100644
--- a/man/DGEList-class.Rd
+++ b/man/DGEList-class.Rd
@@ -8,9 +8,11 @@ A list-based S4 class for storing read counts and associated information from di
}
\section{List Components}{
+For objects of this class, rows correspond to genomic features and columns to samples.
+The genomic features are called genes, but in reality might correspond to transcripts, tags, exons etc.
Objects of this class contain the following essential list components:
\tabular{ll}{
- \code{counts } \tab numeric matrix of read counts, one row for each gene (or genomic feature) and one column for each sample (library).\cr
+ \code{counts } \tab numeric matrix of read counts, one row for each gene and one column for each sample.\cr
\code{samples } \tab data.frame with a row for each sample and columns \code{group}, \code{lib.size} and \code{norm.factors} containing the group labels, library sizes and normalization factors.
Other columns can be optionally added to give more detailed sample information.
@@ -31,7 +33,10 @@ Optional components include:
}
}
-\seealso{ \code{\link{DGEList}} }
+\seealso{
+\code{\link{DGEList}} constructs DGEList objects.
+Other classes defined in edgeR are \code{\link{DGEExact-class}}, \code{\link{DGEGLM-class}}, \code{\link{DGELRT-class}}, \code{\link{TopTags-class}}
+}
\section{Methods}{
This class inherits directly from class \code{list}, so \code{DGEList} objects can be manipulated as if they were ordinary lists.
@@ -39,7 +44,7 @@ However they can also be treated as if they were matrices for the purposes of su
The dimensions, row names and column names of a \code{DGEList} object are defined by those of \code{counts}, see \code{\link{dim.DGEList}} or \code{\link{dimnames.DGEList}}.
-\code{DGEList} objects can be subsetted, see \code{\link{subsetting}}.
+\code{DGEList} objects can be subsetted, see \code{\link[edgeR:subsetting]{subsetting}}.
\code{DGEList} objects also have a \code{show} method so that printing produces a compact summary of their contents.
}
diff --git a/man/camera.DGEList.Rd b/man/camera.DGEList.Rd
index 39ba895..5cfc8d2 100644
--- a/man/camera.DGEList.Rd
+++ b/man/camera.DGEList.Rd
@@ -21,8 +21,8 @@ Test whether a set of genes is highly ranked relative to other genes in terms of
}
\details{
-\code{camera.DGEList} implements methods proposed by Wu and Smyth (2012), but for the digital gene expression data, eg. RNA-Seq data.
-\code{camera.DGEList} performs a \emph{competitive} test in the sense defined by Goeman and Buhlmann (2007).
+This function implements a method proposed by Wu and Smyth (2012) for the digital gene expression data, eg. RNA-Seq data.
+\code{camera} performs a \emph{competitive} test in the sense defined by Goeman and Buhlmann (2007).
It tests whether the genes in the set are highly ranked in terms of differential expression relative to genes not in the set.
It has similar aims to \code{geneSetTest} but accounts for inter-gene correlation.
See \code{\link{roast.DGEList}} for an analogous \emph{self-contained} gene set test.
@@ -31,18 +31,12 @@ The function can be used for any sequencing experiment which can be represented
The design matrix for the experiment is specified as for the \code{\link{glmFit}} function, and the contrast of interest is specified as for the \code{\link{glmLRT}} function.
This allows users to focus on differential expression for any coefficient or contrast in a model by giving the vector of test statistic values.
-\code{camera.DGEList} estimates p-values after adjusting the variance of test statistics by an estimated variance inflation factor.
+\code{camera} estimates p-values after adjusting the variance of test statistics by an estimated variance inflation factor.
The inflation factor depends on estimated genewise correlation and the number of genes in the gene set.
}
\value{
-A data.frame with a row for each set and the following columns:
-\item{NGenes}{number of genes in set}
-\item{Correlation}{inter-gene correlation}
-\item{Direction}{direction of change (\code{"Up"} or \code{"Down"})}
-\item{PValue}{two-tailed p-value}
-\item{FDR}{Benjamini and Hochberg FDR adjusted p-value}
-
+A data.frame. See \code{\link{camera}} for details.
}
\author{Yunshun Chen, Gordon Smyth}
diff --git a/man/cutWithMinN.Rd b/man/cutWithMinN.Rd
index 16b092f..c70d8ae 100644
--- a/man/cutWithMinN.Rd
+++ b/man/cutWithMinN.Rd
@@ -5,15 +5,15 @@
\description{
Discretizes a numeric vector.
-Divides the range of \code{x} into intervals, so that each interval contains a minimum number of values, and codes the values in \code{x} according to which interval they fall.
+Divides the range of \code{x} into intervals, so that each interval contains a minimum number of values, and codes the values in \code{x} according to which interval they fall into.
}
\usage{cutWithMinN(x, intervals=2, min.n=1)}
\arguments{
\item{x}{numeric vector.}
- \item{intervals}{number of intervals (greater than or equal to 2).}
- \item{min.n}{minimum number of values in any interval.}
+ \item{intervals}{number of intervals required.}
+ \item{min.n}{minimum number of values in any interval. Must be greater than \code{length(x)/intervals}.}
}
\value{
@@ -26,6 +26,9 @@ A list with components:
This function strikes a compromise between the base functions \code{cut},
which by default cuts a vector into equal length intervals,
and \code{quantile}, which is suited to finding equally populated intervals.
+It finds a partition of the \code{x} values that is as close as possible to equal length intervals while keeping at least \code{min.n} values in each interval.
+
+Tied values of \code{x} are broken by random jittering, so the partition may vary slightly from run to run if there are many tied values.
}
\author{Gordon Smyth}
@@ -36,7 +39,7 @@ and \code{quantile}, which is suited to finding equally populated intervals.
\examples{
x <- c(1,2,3,4,5,6,7,100)
-cutWithMinN(x,3,min.n=1)
+cutWithMinN(x,intervals=3,min.n=1)
}
\keyword{category}
diff --git a/man/dimnames.Rd b/man/dimnames.Rd
index f3a7621..d89a87c 100644
--- a/man/dimnames.Rd
+++ b/man/dimnames.Rd
@@ -1,13 +1,13 @@
\name{dimnames}
\alias{dimnames.DGEList}
-\alias{dimnames<-.DGEList}
\alias{dimnames.DGEExact}
-\alias{dimnames<-.DGEExact}
\alias{dimnames.DGEGLM}
-\alias{dimnames<-.DGEGLM}
+\alias{dimnames.DGELRT}
\alias{dimnames.TopTags}
-\alias{dimnames<-.TopTags}
-\title{Retrieve the Dimension Names of a DGEList Object}
+\alias{dimnames<-.DGEList}
+\alias{dimnames<-.DGEGLM}
+
+\title{Retrieve the Dimension Names of a DGE Object}
\description{
Retrieve the dimension names of a digital gene expression data object.
}
@@ -16,11 +16,13 @@ Retrieve the dimension names of a digital gene expression data object.
\method{dimnames}{DGEList}(x) <- value
}
\arguments{
- \item{x}{an object of class \code{DGEList}, \code{DGEExact}, \code{DGEGLM} or \code{TopTags}}
- \item{value}{a possible value for \code{dimnames(x)}: see \code{\link{dimnames}}}
+ \item{x}{an object of class \code{DGEList}, \code{DGEExact}, \code{DGEGLM}, \code{DGELRT} or \code{TopTags}}
+ \item{value}{a possible value for \code{dimnames(x)}, see \code{\link{dimnames}}}
}
\details{
-The dimension names of a microarray object are the same as those of the most important matrix component of that object.
+The dimension names of a DGE data object are the same as those of the most important component of that object.
+
+Setting dimension names is currently only permitted for \code{DGEList} or \code{DGEGLM} objects.
A consequence is that \code{rownames} and \code{colnames} will work as expected.
}
@@ -31,7 +33,4 @@ If a list, its components are either \code{NULL} or a character vector the lengt
\author{Gordon Smyth}
\seealso{
\code{\link[base]{dimnames}} in the base package.
-
- \link{02.Classes} gives an overview of data classes used in LIMMA.
}
-\keyword{array}
diff --git a/man/dispBinTrend.Rd b/man/dispBinTrend.Rd
index 618c1f5..499554f 100644
--- a/man/dispBinTrend.Rd
+++ b/man/dispBinTrend.Rd
@@ -9,7 +9,7 @@ Estimate the abundance-dispersion trend by computing the common dispersion for b
\usage{
dispBinTrend(y, design=NULL, offset=NULL, df = 5, span=0.3, min.n=400, method.bin="CoxReid",
- method.trend="spline", trace=0, AveLogCPM=NULL, \dots)
+ method.trend="spline", AveLogCPM=NULL, \dots)
}
\arguments{
@@ -21,7 +21,6 @@ dispBinTrend(y, design=NULL, offset=NULL, df = 5, span=0.3, min.n=400, method.bi
\item{min.n}{minimim number of genes in a bins.}
\item{method.bin}{method used to estimate the dispersion in each bin. Possible values are \code{"CoxReid"}, \code{"Pearson"} or \code{"deviance"}.}
\item{method.trend}{type of curve to smooth the bins. Possible values are \code{"spline"} for a natural cubic regression spline or \code{"loess"} for a linear lowess curve.}
-\item{trace}{logical, should iteration information be output?}
\item{AveLogCPM}{numeric vector giving average log2 counts per million for each gene}
\item{\dots}{other arguments are passed to \code{estimateGLMCommonDisp}}
}
diff --git a/man/glmfit.Rd b/man/glmfit.Rd
index 51a9c23..f283623 100644
--- a/man/glmfit.Rd
+++ b/man/glmfit.Rd
@@ -114,7 +114,7 @@ McCarthy, DJ, Chen, Y, Smyth, GK (2012). Differential expression analysis of mul
Lund, SP, Nettleton, D, McCarthy, DJ, and Smyth, GK (2012).
Detecting differential expression in RNA-sequence data using quasi-likelihood with shrunken dispersion estimates.
\emph{Statistical Applications in Genetics and Molecular Biology} Volume 11, Issue 5, Article 8.
-\url{http://www.statsci.org/smyth/pubs/121013-QuasiSeqPreprint.pdf}
+\url{http://www.statsci.org/smyth/pubs/QuasiSeqPreprint.pdf}
}
\author{Davis McCarthy and Gordon Smyth}
diff --git a/man/plotMDS.DGEList.Rd b/man/plotMDS.DGEList.Rd
index 4476090..2604e8a 100644
--- a/man/plotMDS.DGEList.Rd
+++ b/man/plotMDS.DGEList.Rd
@@ -46,13 +46,7 @@ This function can be slow when there are many libraries.
\value{
A plot is created on the current graphics device.
-An object of class \code{"MDS"} is invisibly returned.
-This is a list containing the following components:
-\item{distance.matrix}{numeric matrix of pairwise distances between columns of \code{x}}
-\item{cmdscale.out}{output from the function \code{cmdscale} given the distance matrix}
-\item{dim.plot}{dimensions plotted}
-\item{x}{x-xordinates of plotted points}
-\item{y}{y-cordinates of plotted points}
+An object of class \code{\link[limma:plotMDS]{MDS}} is invisibly returned.
}
\author{Yunshun Chen, Mark Robinson and Gordon Smyth}
diff --git a/man/roast.DGEList.Rd b/man/roast.DGEList.Rd
index 50217b2..42d7c1e 100644
--- a/man/roast.DGEList.Rd
+++ b/man/roast.DGEList.Rd
@@ -36,26 +36,13 @@ Rotation gene set testing for Negative Binomial generalized linear models.
}
\value{
-\code{roast.DGEList} produces an object of class \code{"Roast"}. This consists of a list with the following components:
- \item{p.value}{data.frame with columns \code{Active.Prop} and \code{P.Value}, giving the proportion of genes in the set contributing meaningfully to significance and estimated p-values, respectively.
-Rows correspond to the alternative hypotheses mixed, up or down.}
- \item{var.prior}{prior value for residual variances.}
- \item{df.prior}{prior degrees of freedom for residual variances.}
-
-
-\code{mroast.DGEList} produces a data.frame with a row for each set and the following columns:
- \item{NGenes}{number of genes in each gene set}
- \item{PropDown}{proportion of genes in set with \code{z < -sqrt(2)}}
- \item{PropUp}{proportion of genes in set with \code{z > sqrt(2)}}
- \item{Direction}{direction of change}
- \item{PValue}{two-sided directional p-value}
- \item{FDR}{directional false discovery rate}
- \item{PValue.Mixed}{non-directional p-value}
- \item{FDR.Mixed}{non-directional false discovery rate}
+\code{roast} produces an object of class \code{\link[limma:roast]{Roast}}. See \code{\link{roast}} for details.
+
+\code{mroast} produces a data.frame. See \code{\link{mroast}} for details.
}
\details{
-This function implements the ROAST gene set test from Wu et al (2010), but for the digital gene expression data, eg. RNA-Seq data.
+This function implements a method for the ROAST gene set test from Wu et al (2010) for the digital gene expression data, eg. RNA-Seq data.
Basically, the Negative Binomial generalized linear models are fitted for count data. The fitted values are converted into z-scores, and then it calls the \code{roast} function in \code{limma} package to conduct the gene set test.
It tests whether any of the genes in the set are differentially expressed.
This allows users to focus on differential expression for any coefficient or contrast in a generalized linear model.
@@ -78,16 +65,16 @@ The output gives p-values three possible alternative hypotheses,
\code{"Down"} to test whether the genes in the set tend to be down-regulated, with negative t-statistics,
and \code{"Mixed"} to test whether the genes in the set tend to be differentially expressed, without regard for direction.
-\code{roast.DGEList} estimates p-values by simulation, specifically by random rotations of the orthogonalized residuals (Langsrud, 2005), so p-values will vary slightly from run to run.
+\code{roast} estimates p-values by simulation, specifically by random rotations of the orthogonalized residuals (Langsrud, 2005), so p-values will vary slightly from run to run.
To get more precise p-values, increase the number of rotations \code{nrot}.
The p-value is computed as \code{(b+1)/(nrot+1)} where \code{b} is the number of rotations giving a more extreme statistic than that observed (Phipson and Smyth, 2010).
This means that the smallest possible p-value is \code{1/(nrot+1)}.
-\code{mroast.DGEList} does roast tests for multiple sets, including adjustment for multiple testing.
-By default, \code{mroast.DGEList} reports ordinary p-values but uses mid-p-values (Routledge, 1994) at the multiple testing stage.
+\code{mroast} does roast tests for multiple sets, including adjustment for multiple testing.
+By default, \code{mroast} reports ordinary p-values but uses mid-p-values (Routledge, 1994) at the multiple testing stage.
Mid-p-values are probably a good choice when using false discovery rates (\code{adjust.method="BH"}) but not when controlling the family-wise type I error rate (\code{adjust.method="holm"}).
-\code{roast.DGEList} performs a \emph{self-contained} test in the sense defined by Goeman and Buhlmann (2007).
+\code{roast} performs a \emph{self-contained} test in the sense defined by Goeman and Buhlmann (2007).
For a \emph{competitive} gene set test, see \code{\link{camera.DGEList}}.
}
diff --git a/man/topTags.Rd b/man/topTags.Rd
index 75e0a84..f9bb409 100755
--- a/man/topTags.Rd
+++ b/man/topTags.Rd
@@ -13,7 +13,7 @@ topTags(object, n=10, adjust.method="BH", sort.by="PValue")
}
\arguments{
-\item{object}{a \code{DGEExact} object (output from \code{exactTest}) or a \code{DGELRT} object (output from \code{glmLRT}), containing the (at least) the elements \code{table}: a data frame containing the log-concentration (i.e. expression level), the log-fold change in expression between the two groups/conditions and the p-value for differential expression, for each tag. If it is a \code{DGEExact} object, then \code{topTags} will also use the \code{comparison} element, which is a vecto [...]
+\item{object}{a \code{\link[edgeR:DGEList-class]{DGEExact}} object (output from \code{exactTest}) or a \code{\link[edgeR:DGELRT-class]{DGELRT}} object (output from \code{glmLRT}), containing the (at least) the elements \code{table}: a data frame containing the log-concentration (i.e. expression level), the log-fold change in expression between the two groups/conditions and the p-value for differential expression, for each tag. If it is a \code{DGEExact} object, then \code{topTags} will a [...]
\item{n}{scalar, number of tags to display/return}
@@ -22,10 +22,19 @@ topTags(object, n=10, adjust.method="BH", sort.by="PValue")
\item{sort.by}{character string, should the top tags be sorted by p-value (\code{"PValue"}), by absolute log-fold change (\code{"logFC"}), or not sorted (\code{"none"}).}
}
-\value{an object of class \code{TopTags} containing the following elements for the top \code{n} most differentially expressed tags as determined by \code{sort.by}.
- \item{table}{a data frame containing the elements \code{logConc}, the log-average concentration/abundance for each tag in the two groups being compared, \code{logFC}, the log-abundance ratio, i.e. fold change, for each tag in the two groups being compared, \code{p.value}, exact p-value for differential expression using the NB model, \code{adj.p.val}, the p-value adjusted for multiple testing as found using \code{p.adjust} using the method specified}
- \item{comparison}{a vector giving the names of the two groups being compared}
-There is a show method for this class.
+\value{
+
+an object of class \code{TopTags} containing the following elements for the top \code{n} most differentially expressed tags as determined by \code{sort.by}:
+
+\item{table}{a data frame containing the elements \code{logFC}, the log-abundance ratio, i.e. fold change, for each tag in the two groups being compared, \code{logCPM}, the log-average concentration/abundance for each tag in the two groups being compared, \code{PValue}, exact p-value for differential expression using the NB model, \code{FDR}, the p-value adjusted for multiple testing as found using \code{p.adjust} using the method specified.}
+\item{adjust.method}{character string stating the method used to adjust p-values for multiple testing.}
+\item{comparison}{a vector giving the names of the two groups being compared.}
+\item{test}{character string stating the name of the test.}
+
+The dimensions, row names and column names of a \code{TopTags} object are defined by those of \code{table}, see \code{\link{dim.TopTags}} or \code{\link{dimnames.TopTags}}.
+
+\code{TopTags} objects also have a \code{show} method so that printing produces a compact summary of their contents.
+
}
\author{Mark Robinson, Davis McCarthy, Gordon Smyth}
diff --git a/src/core/glm.h b/src/core/glm.h
index edf9b45..7f8569d 100644
--- a/src/core/glm.h
+++ b/src/core/glm.h
@@ -34,6 +34,7 @@ private:
bool failed;
double nb_deviance(const double*, const double*, const double&) const;
+ void autofill(const double*, double*, const double*);
};
class adj_coxreid {
--
Alioth's /git/debian-med/git-commit-notice on /srv/git.debian.org/git/debian-med/r-bioc-edger.git
More information about the debian-med-commit
mailing list