A function to get the Red-Blue color scheme used by dChip

Description

A simple, vectorized function that computes a Red/Blue color for plotting microarray expression data.

Usage

GetColor(value, GreenRed=FALSE, DisplayRange=3)
dChip.colors(n)
greenred.colors(n)

Arguments

Details

GetColor is a simple mapping into RGB land provided by Cheng Li. dChip.colors provides functionality similar to that of topo.colors for the red--blue colors used for genome plots. greenred.colors does the same for the green-black-red gradient.

Value

A vector of RGB colors suitable for plotting in R.

Author

R. Gentleman, based on an original by C. Li.

Examples

set.seed(10)
x <- rnorm(10)
GetColor(x)
dChip.colors(10)

Produce Smoothed Sense/Anti-sense For All Chromosomes

Description

'Makesense' takes either an ExpressionSet object or a matrix of gene expressions and will produce a smoothed positive and negative strands for all chromosomes.

Usage

Makesense(expr, lib, ...)

Arguments

Details

The expr argument can either be of class ExpressionSet or matrix , where the latter represents the matrix of gene expressions.

If the expr argument is an ExpressionSet , the lib argument will use the annotation slot. Users can override this behaviour and supply their own lib argument if they wish. If the ExpressionSet has no value associated with the annotation slot (which should not happen, but is possible) then the user must supply the lib argument manually or the function will throw an error.

Value

A list of 2 components:

*

Seealso

plotChr

Author

Robert Gentleman and Xiaochun Li

Examples

if (require("hgu133a.db")) {
data(expressionSet133a)
esetobj <- Makesense(exprs(expressionSet133a), "hgu133a")
esetobj2 <- Makesense(expressionSet133a[1:200, ])
}

A function for plotting expression data from an ExpressionSet for a given chromosome.

Description

Given a particular ExpressionSet object, a chromLocation object, and a chromosome name, will plot selected ExpressionSet data using various methods.

Usage

alongChrom(eSet, chrom, specChrom,  xlim, whichGenes, 
   plotFormat=c("cumulative", "local","image"),
   xloc=c("equispaced", "physical"), 
   scale=c("none","zscale","rankscale","rangescale","zrobustscale"),
   geneSymbols=FALSE, byStrand=FALSE, colors="red", lty=1, type="S",
   ...)

Arguments

Details

The genes on the chromosome of interest are extracted from the chromLocation object passed in, which are then intersected with the genes listed in the ExpressionSet. These remaining genes will then be plotted according to the plotFormat argument. If image is specified, an image plot is created showing the expression levels of the samples by gene, using a colour map to denote the levels. If cumulative is chosen, the cumulative expression level is plotted against the genes for each sample. Likewise, if local is used, the raw data is plotted for each sample against the genes using a boxplot format.

Not all parameters are honored for all plotformats. xloc , lty , and type are only used with the cumulative plotformat.

Author

Jeff Gentry

Examples

data(sample.ExpressionSet)
## A bit of a hack to not have a package dependency on hgu95av2
## but need to fiddle w/ the warn level to not fail the example anyways.
curWarn <- options(warn=0)
on.exit(options(curWarn), add=TRUE)
if (require("hgu95av2.db")) {
z <- buildChromLocation("hgu95av2")
lty <- c(1, 2, 3, 4, 5)
cols <- c("red", "green", "blue", "orange", "magenta", "black")
cols <- cols[sample.ExpressionSet$type]
if (interactive()) {
par(ask=TRUE)
}

## Here we're using xlim to denote a physical region to display
xlim <- c(87511280,127717880)
for (xl in c("equispaced", "physical"))
for (sc in c("none","rangescale"))
{
alongChrom(sample.ExpressionSet, "1", z, xlim=xlim, xloc=xl,
plotFormat="cumulative", scale=sc,lty=lty, colors=cols)
}

## Here we're looking for specific genes
which <- c("31540_at","31583_at", "31508_at", "31529_at", "31439_f_at",
"31729_at")
## Gene "31529_at" does not exist in the current set of genes,
## here it demonstrates how genes not available are dropped.
for (xl in c("equispaced", "physical"))
for (sc in c("none","rangescale"))
{
alongChrom(sample.ExpressionSet, "1", z, which=which, xloc=xl,
plotFormat="cumulative", scale=sc,lty=lty, col=cols)
}

## Do an image plot
for (bs in c(TRUE,FALSE))
alongChrom(sample.ExpressionSet, "1",z, xlim=xlim, plotFormat="image",
scale="zscale", byStrand=bs)

## A boxplot
for (st in c(TRUE,FALSE))
alongChrom(sample.ExpressionSet, "1", z, plotFormat="local",
colors=cols, byStrand=st)
} else print("Example can not be run without the hgu95av2 data package")

Create an amplicon plot

Description

Given a two-sample test statistic and an ExpressionSet this function plots regions of the genome that are either highly expressed (in red) or have low expression (blue) differentially in the two groups.

Usage

amplicon.plot(ESET, FUN, genome)

Arguments

Details

In some genetic studies we are interested in finding regions of the genome where there are a set of highly expressed genes in some subgroup of the population. This set of highly (or lowly) expressed genes is often of great interest. For example in breast cancer the HER--2 gene is on an amplicon. In some patients approximately 5 genes located near HER--2 are all amplified.

These plot should help in the search for such regions.

Value

No value is returned. This function is executed purely for side effect.

Seealso

esApply , make.chromOrd

Author

Robert Gentleman

Examples

##none yet; takes too long

A function for marking specific probes on a cPlot.

Description

Given a set of probes, will highlight them in the color desired on a plot which has already been created via the function cPlot().

Usage

cColor(probes, color, plotChroms, scale=c("relative","max"), glen=0.4,
       ...)

Arguments

Details

It is important to call the function cPlot() first. This function will then search for the specific locations of the probes desired, which are contained within the plotChroms instance of a chromLocation class. It will then pass these on to the plotting routine to highlight the desired locations. NOTE: It is important that plotChroms , scale and glen parameters are the same as used for cPlot() .

Seealso

cPlot , chromLocation-class

Author

Jeff Gentry

Examples

if (require("hgu95av2.db")) {
z <- buildChromLocation("hgu95av2")
cPlot(z)
probes <- c("266_s_at", "31411_at", "610_at", "failExample")
cColor(probes, "red", z)
probes2 <- c("960_g_at", "41807_at", "931_at", "39032_at")
cColor(probes2, "blue", z)
} else
print("Need hgu95av2.db data package for the example")

A plotting function for chromosomes.

Description

Given a chromLocation object, will plot all the gene locations from that object.

Usage

cPlot(plotChroms, useChroms=chromNames(plotChroms),
      scale=c("relative","max"), fg="white", bg="lightgrey", 
      glen=0.4, xlab="", ylab="Chromosome",
      main = organism(plotChroms), ...)

Arguments

Details

This function will first use the lengths of the chromosomes, stored in the object to create scaling factors for the X axis. Once the scaling factors are determined, the chromLocation object which is passed in is used to determine all the gene locations/strand information/etc, which is then plotted for the user.

Seealso

cScale , cColor , chromLocation-class

Author

Jeff Gentry

Examples

## A bit of a hack to not have a package dependency on hgu95av2
## but need to fiddle w/ the warn level to not fail the example anyways.

curWarn <- options(warn=0)
on.exit(options(curWarn), add=TRUE)
if (require("hgu95av2.db")) {
z <- buildChromLocation("hgu95av2")

if (interactive()) {
curPar <- par(ask=TRUE)
on.exit(par(curPar), add=TRUE)
}

for (sc in c("max","relative"))
cPlot(z,c("1","5","10","X","Y"),sc)
} else print("This example can not be run without hgu95av2 data package")

A function for mapping chromosome length to a number of points.

Description

Given a number of points (generally representing the number of points on a plot's axis), and a vector of chromosome lengths - will generate a vector of the same length as the one passed in containing scaling factors for each chromosome.

Usage

cScale(points, cLengths, method=c("max", "relative"), chrom)

Arguments

Details

The scale factor is calculated in a manner based on the method argument. If method is max , the factor is derived by dividing the points argument by each chromosome's length (in base pairs). If the method chosen is relative , then the scale is determined by dividing the points argument by the maximum chromsome length, and applying that value to each chromosome.

Seealso

cPlot

Author

Jeff Gentry

Examples

## A bit of a hack to not have a package dependency on hgu95av2
## but need to fiddle w/ the warn level to not fail the example anyways.
curWarn <- options(warn=0)
on.exit(options(warn), add=TRUE)
if (require("hgu95av2.db")) {
z <- buildChromLocation("hgu95av2")

for (sc in c("max","relative"))
scale <- cScale(1000, chromLengths(z),sc,"Y")
} else print("This example needs the hgu95av2 data package")

A small dataset for testing

Description

An artificial Affymetrix hgu133a dataset, with one covariate 'cov1'.

Format

The data are artifical. There are 6 cases labeled 1 to 6 and and 22283 genes as in an Affymetrix U133a chips. There is one covariate (factor) whose values are "type 1" for the first 3 samples and "type 2" for the last 3 samples.

Usage

data(expressionSet133a)

Examples

data(expressionSet133a)

Heatmap of a matrix with grouped rows and columns

Description

The function uses grid.rect and grid.rect to draw a heatmap with grouped rows and columns.

Usage

groupedHeatmap(z, frow, fcol,
  fillcolours = c("#2166ac","#4393c3","#92c5de","#d1e5f0","#fefefe","#fddbc7","#f4a582","#d6604d","#b2182b"),
  bordercolour = "#e0e0e0",
  zlim = range(z, na.rm=TRUE))

Arguments

Details

The function can be called within other drawing operations from the grid package, e.g. within a viewport.

Value

The function is called for its side effect, drawing text and rectangles on the current viewport.

Seealso

grid.text , grid.rect

Author

Wolfgang Huber http://www.ebi.ac.uk/huber

Examples

data("mtcars")

groupedHeatmap(
scale(mtcars),
frow = factor(sapply(strsplit(rownames(mtcars), " "), "[", 1)),
fcol = factor(round(seq_len(ncol(mtcars))/3)))

Stacked histogram

Description

Stacked histogram

Usage

histStack(x, breaks, breaksFun=paste, ylab="frequency", ...)

Arguments

Details

The function calls hist for each element of x and plots the frequencies as a stacked barplot using barplot with beside=FALSE .

Value

The function is called for its side effect, producing a barplot on the active graphics device. It returns the result of the call to barplot .

Author

Wolfgang Huber http://www.ebi.ac.uk/huber

Examples

x    <- list(rnorm(42), rnorm(42)+2)
br   <- seq(-3, 5, length=13)
cols <- c("#1D267B", "#ceffc0")
histStack(x, breaks=br, col=cols)

histStack(x, breaks=br, col=cols,
breaksFun=function(z) paste(signif(z, 3)))

Write an HTML IMG tag together with a MAP image map.

Description

Write an HTML IMG tag together with a MAP image map.

Usage

list(list("imageMap"), list("matrix,connection,list,character"))(object, con, tags, imgname)

Arguments

Details

The most important tags are TITLE , HREF , and TARGET . If the list tags contains an element with name TITLE , then this must be a named character vector containing the tooltips that are to be displayed when the mouse moves over a node. The names of the nodes are specified in the names attribute of the character vector and must match those of object .

Similarly, HREF may be used to specify hyperlinks that the browser can follow when the mouse clicks on a node, and TARGET to specify the target browser window.

Currently, only rectangular regions are implemented; the actual shape of the nodes as specified in object is ignored. Also, tags for edges of the graph are currently not supported.

This function is typically used with the following sequence of steps:

• generate your graphic and save it as a bitmap file, e.g. using the jpeg , png , or bitmap device. At this stage, you also need to figure out the pixel coordinates of the interesting regions within your graphic. Since the mapping between device coordinates and pixel coordinates is not obvious, this may be a little tricky. See the examples below, and for a more complex example, see the source code of the function plotPlate .

• open an HTML page for writing and write HTML header, e.g. using the openHtmlPage function.

• Call the imageMap function.

• Optionally, write further text into the HTML connection.

• Close HTML file, e.g. using the closeHtmlPage function.

Value

The function is called for its side effect, which is writing text into the connection con .

Seealso

plotPlate , writeLines

Author

Wolfgang Huber http://www.dkfz.de/abt0840/whuber

Examples

f1  = paste(tempfile(), ".html", sep="")
f2  = paste(tempfile(), ".html", sep="")
fpng = tempfile()

if(capabilities()["png"]) {
## create the image
colors = c("#E41A1C","#377EB8","#4DAF4A","#984EA3","#FF7F00","#FFFF33","#A65628","#F781BF","#999999")
width  = 512
height = 256
png(fpng, width=width, height=height)
par(mai=rep(0,4))
plot(0,xlim=c(0,width-1),ylim=c(0,height-1),xaxs="i",yaxs="i",type="n",bty="n")
cx=floor(runif(100)*(width-11))
cy=floor(runif(100)*(height-11))
coord=cbind(cx, cy, cx+10, cy+10)
rect(coord[,1], height-coord[,2], coord[,3], height-coord[,4],
col=sample(colors, 100, replace=TRUE))
text(width/2, height-3, "Klick me!", adj=c(0.5, 1), font=2)
dev.off()

## create the frame set
cat("<html><head><title>Hello world</title></head>
",
"<frameset rows="280,*" border="0">
",
"<frame name="banner" src="file://", f2, "">
",
"<frame name="main" scrolling="auto">",
"</frameset>", sep="",file=f1)

## create the image map
href  =sample(c("www.bioconductor.org", "www.r-project.org"),nrow(coord),replace=TRUE)
title =sample(as.character(packageDescription("geneplotter")),nrow(coord),replace=TRUE)
con = file(f2, open="w")
imageMap(coord, con,
list(HREF=paste("http://", href, sep=""),
TITLE=title, TARGET=rep("main", nrow(coord))), fpng)
close(con)

cat("Now have a look at file ", f1, " with your browser.
", sep="")
}

Make a chromOrd object

Description

This function makes a chromOrd object.

Usage

make.chromOrd(genome, gnames)

Arguments

Details

This function reads in a lot of annotation data and creates a list with one element for each chromosome. The elements of this list are indices indicating the order of the genes that are on that chromosome (and in the annotation data set being used).

Value

A list of chromOrd type. One element for each chromosome. Suitable for reordering other values according to the chromosomal location.

Seealso

amplicon.plot

Author

Robert Gentleman

Examples

data(sample.ExpressionSet)
make.chromOrd("hgu95A", featureNames(sample.ExpressionSet))

Multiple empirical cumulative distribution functions (ecdf) and densities

Description

Plot multiple empirical cumulative distribution functions (ecdf) and densities with a user interface similar to that of boxplot . The usefulness of multidensity is variable, depending on the data and the smoothing kernel. multiecdf will in many cases be preferable. Please see Details.

Usage

multiecdf(x, list())
list(list("multiecdf"), list("formula"))(formula, data = NULL, xlab, na.action = NULL, list())
list(list("multiecdf"), list("matrix"))(x, xlab, ...) 
list(list("multiecdf"), list("list"))(x,
          xlim,
          col = brewer.pal(9, "Set1"),
          main = "ecdf",
          xlab,
          do.points = FALSE,
          subsample = 1000L,
          legend = list(
            x = "right",
            legend = if(is.null(names(x))) paste(seq(along=x)) else names(x),
            fill = col),
          list())
multidensity(x, list())
list(list("multidensity"), list("formula"))(formula, data = NULL, xlab, na.action = NULL, list())
list(list("multidensity"), list("matrix"))(x, xlab, ...) 
list(list("multidensity"), list("list"))(x,
             bw = "nrd0",
             xlim,
             ylim,
             col  = brewer.pal(9, "Set1"),
             main = if(length(x)==1) "density" else "densities",
             xlab,
             lty  = 1L,
             legend = list(
               x = "topright",
               legend = if(is.null(names(x))) paste(seq(along=x)) else names(x),
               fill = col),
             density = NULL,
             list())

Arguments

Details

Density estimates : multidensity uses the function density . If the density of the data-generating process is smooth on the real axis, then the output from this function tends to produce results that are good approximations of the true density. If, however, the true density has steps (this is in particular the case for quantities such as p-values and correlation coefficients, or for some distributions that have weight only on the posititve numbers, or only on integer numbers), then the output of this function tends to be misleading. In that case, please either use multiecdf or histograms, or try to improve the density estimate by setting the density argument ( from , to , kernel ).

Bandwidths : the choice of the smoothing bandwidths in multidensity can be problematic, in particular, if the different groups vary with respect to range and/or number of data points. If curves look excessively wiggly or overly smooth, try varying the arguments xlim and bw ; note that the argument bw can be a vector, in which case it is expect to align with the groups.

Value

For the multidensity functions, a list of density objects.

Seealso

boxplot , ecdf , density

Author

Wolfgang Huber

Examples

words = strsplit(packageDescription("geneplotter")$Description, " ")[[1]]
factr = factor(sample(words, 2000, replace = TRUE))
x = rnorm(length(factr), mean=as.integer(factr))

multiecdf(x ~ factr)
multidensity(x ~ factr)

Open and close an HTML file for writing.

Description

Open and close an HTML file for writing..

Usage

openHtmlPage(name, title="")
  closeHtmlPage(con)

Arguments

Details

See example.

Value

For openHtmlPage , a connections .

Author

Wolfgang Huber http://www.dkfz.de/abt0840/whuber

Examples

fn <- tempfile()
con <- openHtmlPage(fn, "My page")
writeLines("Hello world", con)
closeHtmlPage(con)
readLines(paste(fn, ".html", sep=""))

Plot Smoothed Sense/Anti-sense of Specified Chromosomes

Description

For a given chromosome, plot the smooths of the sense and the anti-sense from 5' to 3' (left to right on x-axis).

Usage

plotChr(chrN, senseObj, cols = rep("black", length(senseObj[[1]])), log = FALSE, xloc = c("equispaced", "physical"), geneSymbols = FALSE, ngenes = 20, lines.at = NULL, lines.col = "red")

Arguments

Seealso

Makesense

Author

Robert Gentleman and Xiaochun Li

Examples

example(Makesense)

if (interactive())
op <- par(ask=TRUE)

cols <- ifelse(expressionSet133a$cov1=="test 1", "red", "green")
plotChr("21", esetobj, cols)

# plot on log-scale:

plotChr("21", esetobj, cols, log=TRUE)

# genesymbol instead of probe names:

plotChr("21", esetobj, cols, log=TRUE, geneSymbols=TRUE)

# add vertical lines at genes of interest:

gs <- c("220372_at", "35776_at", "200943_at")
plotChr("21", esetobj, cols, log=TRUE, geneSymbols=FALSE, lines.at=gs)

# add vertical lines at genes of interest
# with specified colors:

gs <- c("220372_at", "35776_at", "200943_at")
cc <- c("blue", "cyan","magenta")
plotChr("21", esetobj, cols, log=TRUE, geneSymbols=FALSE, lines.at=gs,
lines.col=cc)
if (interactive())
par(op)

A function to plot a graph colored by expression data

Description

Given a graph and expression data for one entity, will plot the graph with the nodes colored according to the expression levels provided.

Usage

plotExpressionGraph(graph, nodeEGmap, exprs, ENTREZIDenvir, mapFun, log = FALSE, nodeAttrs = list(), ...)

Arguments

Details

This function can be used to plot a graph and have the nodes colored according to expression levels provided by the user. The graph parameter is a graph object from the graph package.

The nodeEGmap parameter is a list that maps the nodes of the graphs to EntrezLink IDs. An example of this is the IMCAEntrezLink object in the integrinMediatedCellAdhesion data set in the graph package.

The exprs argument is a vector mapping expression levels to Affymetrix IDs. One way to generate an appropriate vector is to extract a single column from an ExpressionSet .

The ENTREZIDenvir environment maps Affymetrix IDs to EntrezLink IDs. The simplest way to provide this argument is to load the preferred Bioconductor data package (e.g. hgu95av2.db ) and pass in that package's xxx2ENTREZID , where xxx is the name of the package.

The mapFun function defaults to the function defMapFun , which maps nodes to be either blue, green or red depending for expression ranges of 0-100, 101-500, and 501+. In the case where log is TRUE these ranges are modified with log2 . Custom versions of this function can be supplied by the user - it must take two parameters, first the expression vector and a boolean value ( log ) specifying if the data has had a log2 applied to it. The function must return a vector with the same names as the expression vector, but the values of the vector will be color strings.

The nodeAttrs list can be specified if any other node attributes are desired to be set by the user. Please see the plot.graph man page for more information on this. The other attribute list ( attrs and edgeAttrs ) can be passed in via the ... parameter.

The IMCAEntrezLink data structure was created for the purpose of illustrating this program. On Sept 24 2007, the current version of hgu95av2.db was used to map from the nodes of IMCAGraph (in graph package) to Entrez identifiers.

Seealso

plot.graph , integrinMediatedCellAdhesion

Author

Jeff Gentry

Examples

if (require("Rgraphviz") && require("hgu95av2.db") &&
require("fibroEset")) {
data(integrinMediatedCellAdhesion)
data(IMCAEntrezLink)
data(fibroEset)
attrs <- getDefaultAttrs()
attrs$graph$rankdir <- "LR"
plotExpressionGraph(IMCAGraph, IMCAEntrezLink,
exprs(fibroEset)[,1],
hgu95av2ENTREZID, attrs = attrs)
}

Generate an MA plot

Description

Generate a plot of log fold change versus mean expression (MA plot)

Usage

list(list("plotMA"), list("data.frame"))( object, ylim = NULL,
  colNonSig = "gray32", colSig = "red3", colLine = "#ff000080",
  log = "x", cex=0.45, xlab="mean expression", ylab="log fold change", ... )

Arguments

Examples

plotMA(
data.frame(
`M` = exp(rexp(1000)),
`A` = rnorm(1000) -> tmp,
`isde` = abs(tmp)>2)
)

Save the contents of the current graphics device to a file

Description

Save the contents of the current graphics device to file

Usage

savepdf(fn, dir, width=6, asp=1)
  saveeps(fn, dir, width=6, asp=1) 
  savepng(fn, dir, width=480, asp=1)
  savetiff(fn, dir, density=360, keeppdf=TRUE, ...)

Arguments

Details

The functions are called for their side effect, writing a graphics file.

savepdf , savepng , and saveeps use the devices pdf , png , and postscript , respectively.

There is currently no TIFF device for R, so savetiff works differently. It relies on the external tool convert from the ImageMagick software package. First, savetiff produces a PDF files with savepdf , then uses system to invoke convert with the parameter density . savetiff does not check for the existence of convert or the success of the system call, and returns silently no matter what.

Value

Character: name of the file that was written.

Seealso

dev.copy , pdf , png , postscript

Author

Wolfgang Huber http://www.dkfz.de/abt0840/whuber

Examples

x = seq(0, 20*pi, len=1000)
plot(x*sin(x), x*cos(x), type="l")

try({   ## on some machines, some of the devices may not be available
c(
savepdf("spiral", dir=tempdir()),
savepng("spiral", dir=tempdir()),
saveeps("spiral", dir=tempdir()),
savetiff("spiral", dir=tempdir())
)
})