In dieterich-lab/CellPlot: Segmented Bar Chart Plot Methods
Description
CellPlot is a GNU R package with plot methods for the
integrated visualisation of functional term enrichment and expression data.
It can be used to display commonly used gene ontology (GO) term enrichment
analysis from differential expression of genes (DEG) studies.
The development platform of this software is hosted on github.
In case of any questions, consider using the issues system.
Usage
Load package and data
After installing the package (see Installation section), load it in a running
R session. Example data can be accessed with the data() function.
library(CellPlot)
data("leukemiasGO")
Plot methods
Example to create a subset of the leukemiasGO data set tables:
x <- subset(leukemiasGO$CLL, pvalCutOff <= 0.05 & Significant > 20)
x <- x[order(-x$LogEnrich),]
Cell Plot
The function cell.plot plots a horizontal barchart of strictly positive values in x.
For each entry, a vector of additional values needs to be provided in a list.
The additional values are plotted as cells subdividing the length of the corresponding bar.
A typical usage scenario is plotting the enrichment of Gene Ontology terms,
with individual cells reflecting the differential regulation of the constituent genes.
cell.plot(x = setNames(x$LogEnrich, x$Term),
cells = x$log2FoldChange,
main ="GO enrichment (NoT vs CLL)",
x.mar = c(.4, 0),
key.n = 7,
y.mar = c(.1, 0),
cex = 1.6,
cell.outer = 3,
bar.scale = .7,
space = .2)
Sym Plot
The function sym.plot plots a split barchart, showing the proportions of
two mutually exclusive sets in relation to a set containing them both.
E.g., Gene Ontology terms, showing the proportions of differentially down-regulated
and up-regulated annotated genes from a perturbation experiment.
The color of the central column elements maps to the value provided in x (e.g. GO term enrichment).
Associated genes may be provided as a list of vectors of expression values,
same as for cell.plot(), or as separate vectors x.up and x.down,
providing the numbers of up- and down-regulated genes in the same order as x.
sym.plot(x = setNames(x$LogEnrich, x$Term),
cells = x$log2FoldChange,
x.annotated = x$Annotated,
main = "GO enrichment (NoT vs CLL)",
x.mar = c(.47, 0),
key.n = 7,
cex = 1.6,
axis.cex = .8,
group.cex = .7)
Arc Plot
An arc.plot represents enrichment bargraphs, showing abundance of positive and negative
values (e.g. log2 fold change of gene expression comparisons). Units (e.g. gene names) are
clustered and intersections are indicated by an arc.
x$up <- lapply(Map(setNames, x$log2FoldChange, x$GenesSignificant), function (i) { i[i>0] })
x$dwn <- lapply(Map(setNames, x$log2FoldChange, x$GenesSignificant), function (i) { i[i<0] })
arc.plot(x = setNames(x$LogEnrich, x$Term),
up.list = x$up,
down.list = x$dwn,
x.mar = c(.9, .5))
Go Histogram
A bargraph showing positive and negative value abundances per category.
Supply a list of data.frames and compare each table, performing a one
sided t-test on the absolute logfc.term values.
Data is prepared as follows:
y <- lapply(leukemiasGO, function (x) {
x$Upregulated <- sapply(x$log2FoldChange, function (z) sum(z>0))
x$Downregulated <- sapply(x$log2FoldChange, function (z) sum(z<0))
x
})
yterms <- unique(unlist(lapply(y, function(x){
x <- subset(x, pvalCutOff <= 0.05)
x <- x[order(x$LogEnrich),]
head(x, 9)$GO.ID
})))
An the plot generated:
par(mar = c(0,.5,2.5,8))
go.histogram(y, go.alpha.term = "pvalCutOff", gene.alpha.term = "padj",
min.genes = 5, max.genes = 1e10, go.selection = yterms, show.ttest = T,
main = "GO enrichment\nin leukemia differential gene expression\ncompared to control samples",
axis.cex = 1, lab.cex = 1.5, main.cex = 1.5)
Visual parameters
Both functions provide a range of parameters to tweak the visual appearance of the plots.
- Font sizes of the various elements can be adjusted using the
cex parameters.
- Internal margins are adjusted through
x.mar and y.mar. Used in particular to accomodate the label text to the left-hand side.
- Absolute scaling can be achieved through the
bar.scale parameter, which acts as a multiplier for a predetermined average bar height. This is meant to be used in conjunction with a graphics device of constant size, to ensure visual consistency among multiple plots that vary in the number of terms displayed.
- Data ranges for all output may be fixed:
elem.bounds Require the cardinality of a term or significant subset to be in a specific range, e.g. c(10,100) -- between 10 and 100 genes.x.bound Upper limit of the value displayed on the x-axis. For sym.plot this is must be a value between 0 and 100. The lower limit is always 0.cell.bounds (only cell.plot) Sets the colour scale for cell data to a fixed range. If the plot contains values outside of that range, an indicator is added to the colour legend. This is particularly useful in the event of a few extreme outliers.mid.bounds (only sym.plot) Similar to cell.bounds, this sets the colour range of the central column cells.sym (only cell.plot), when set to TRUE, makes the scale of cell values symmetrical.
- Colour schemes may be controlled through character vectors provided to the following parameters. Internally, the R native function
ColorRampPalette() is used.
cell.col (only cell.plot) Specify three valid colour names for cell data visualisation (lower extreme, mid point, upper extreme).mid.col (only sym.plot) Specify two valid colour names to define the range of the central column cell colours.
- Colour keys can be adjusted in their resolution (number of boxes) using the
key.n parameter.
- Gridlines may be hidden via the
gridlines parameter.
Workflows
Gene ontology term enrichment in differential gene expression data can be
performed using the
Bioconductor/topGO
package.
Dataset golubGO
This section provides code and comments on the workflow to perform differential
expression of genes (DEG) analysis and the gene ontology term (GO) enrichment
of from the results applied to the golub dataset from the
Bioconductor/leuke
package.
Dataset leukemiasGO
This section provides code and comments on the workflow to perform differential
expression of genes analysis and the gene ontology term enrichment
of from the results applied to the leukemiasEset dataset from the
Bioconductor/leukemiasEset package.
GO annotation was performed with the
Bioconductor/annotate
# CRAN
library(parallel)
library(dplyr)
library(stringr)
# Bioconductor
library(BiocParallel)
library(DESeq2)
library(topGO)
library(annotate)
library(leukemiasEset)
data(leukemiasEset)
# M <- select(hu6800.db, featureNames(leukemiasEset), c("ENSEMBL","GO"), keytype = "ENSEMBL")
# M <- subset(M, ONTOLOGY == "BP", c("ENSEMBL","GO"))
# M <- M[!duplicated(M),]
# M <- dlply(M, "ENSEMBL", function (x) unique(x$GO))
# M <- as.character(unique(unlist(select(org.Hs.eg.db,featureNames(leukemiasEset),"ENSEMBL",keytype = "ENSEMBL"))))
A <- as(leukemiasEset,"data.frame")
A <- subset(A, LeukemiaType %in% c("NoL","ALL", "AML", "CLL"))
As <- subset(A, select = "LeukemiaType")
As$LeukemiaType <- relevel(factor(as.character(As$LeukemiaType)), "NoL")
A <- subset(A, select = grepl("^ENS", colnames(A)))
A <- sapply(A, as.integer)
A <- t(A)
DEG <- DESeqDataSetFromMatrix(countData = A, colData = As, design = ~ LeukemiaType)
DEG <- DESeq(DEG, fitType = "mean", parallel = T, BPPARAM = MulticoreParam(20))
n <- levels(As$LeukemiaType)[-1]
leukemiasGO <- lapply(setNames(n, n), function (n) {
x <- results(DEG, c("LeukemiaType", n, "NoL"), "LeukemiaType", alpha = .05)
x <- as.data.frame(x)
x$padj[is.na(x$padj)] <- 1
g <- new("topGOdata", ontology = "BP", description = 'Leukemia',
allGenes = setNames(x$padj, rownames(x)),
mapping = "org.Hs.eg.db",
geneSelectionFun = function (allScore) { allScore <= 0.05 },
annotationFun = annFUN.org, ID = "Ensembl")
t <- new("elimCount", testStatistic = GOFisherTest, name = "Fisher test") # test definition
s <- getSigGroups(g, t) # run F-test
r <- GenTable(g, pvalCutOff = s, topNodes = length(g@graph@nodes)) # return data.frame
r$pvalCutOff <- as.numeric(str_replace_all(r$pvalCutOff, "[^0-9e\\-\\.]*", ""))
r$LogEnrich <- log2(r$Significant / r$Expected)
ga <- genesInTerm(g) # GenesAnnotated | list of genes per go-terms
ga <- ga[r$GO.ID] # eliminate missing terms
names(ga) <- NULL
r$GenesAnnotated <- ga
xs <- x[,c("padj", "log2FoldChange")] # significant stats subset
xs <- subset(xs, padj < 0.05)
r$GenesSignificant <- lapply(r$GenesAnnotated, intersect, rownames(xs)) # extract genes
ei.rows <- mclapply(r$GenesSignificant, function (y) {
if (length(y)) as.list(xs[y,,drop=FALSE])
else as.list(rep(NA_real_, length(xs)))
}, mc.cores = 10)
ei <- mclapply(names(xs), function(z) {
lapply(ei.rows, "[[", z)
}, mc.cores = 10)
ei <- structure(ei, names = names(xs), row.names = seq(nrow(r)), class = "data.frame")
row.names(ei) <- NULL
r <- data.frame(r, ei, stringsAsFactors = FALSE, check.names = FALSE)
return(r)
})
# lapply(GO, function(x) {list(all(r$Annotated == sapply(r$GenesAnnotated, length)),
# all(r$Significant == sapply(r$GenesSignificant, length)))})
leukemiasGO <- lapply(leukemiasGO, function(x) subset(x, pvalCutOff < 0.1))
save(leukemiasGO, file = "data/leukemiasGO.rdata")
Installation
CellPlot
Install the latest version from the repository on
github.com/dieterich-lab/CellPlot.
With the devtools package, it is an easy task:
library(devtools)
install_github('dieterich-lab/CellPlot', build_vignettes = TRUE)
Dependencies
To install the packages from Bioconducter, use the following steps:
# load bioconductor functions
source("https://bioconductor.org/biocLite.R")
# install packages
biocLite("BiocParallel")
biocLite("annotate")
biocLite("DESeq2")
biocLite("topGO")
biocLite("leukemiasEset")
Dependencies from CRAN can be installed via:
# install packages
install.packages(c("stringr","dplyr","knitr","rmarkdown","devtools"))
Session info
# session info from build machine
sessionInfo()
dieterich-lab/CellPlot documentation built on May 15, 2019, 8:29 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description
CellPlot is a GNU R package with plot methods for the integrated visualisation of functional term enrichment and expression data. It can be used to display commonly used gene ontology (GO) term enrichment analysis from differential expression of genes (DEG) studies.
The development platform of this software is hosted on github. In case of any questions, consider using the issues system.
Usage
Load package and data
After installing the package (see Installation section), load it in a running
R session. Example data can be accessed with the data() function.
library(CellPlot) data("leukemiasGO")
Plot methods
Example to create a subset of the leukemiasGO data set tables:
x <- subset(leukemiasGO$CLL, pvalCutOff <= 0.05 & Significant > 20) x <- x[order(-x$LogEnrich),]
Cell Plot
The function cell.plot plots a horizontal barchart of strictly positive values in x.
For each entry, a vector of additional values needs to be provided in a list.
The additional values are plotted as cells subdividing the length of the corresponding bar.
A typical usage scenario is plotting the enrichment of Gene Ontology terms,
with individual cells reflecting the differential regulation of the constituent genes.
cell.plot(x = setNames(x$LogEnrich, x$Term), cells = x$log2FoldChange, main ="GO enrichment (NoT vs CLL)", x.mar = c(.4, 0), key.n = 7, y.mar = c(.1, 0), cex = 1.6, cell.outer = 3, bar.scale = .7, space = .2)
Sym Plot
The function sym.plot plots a split barchart, showing the proportions of
two mutually exclusive sets in relation to a set containing them both.
E.g., Gene Ontology terms, showing the proportions of differentially down-regulated
and up-regulated annotated genes from a perturbation experiment.
The color of the central column elements maps to the value provided in x (e.g. GO term enrichment).
Associated genes may be provided as a list of vectors of expression values,
same as for cell.plot(), or as separate vectors x.up and x.down,
providing the numbers of up- and down-regulated genes in the same order as x.
sym.plot(x = setNames(x$LogEnrich, x$Term), cells = x$log2FoldChange, x.annotated = x$Annotated, main = "GO enrichment (NoT vs CLL)", x.mar = c(.47, 0), key.n = 7, cex = 1.6, axis.cex = .8, group.cex = .7)
Arc Plot
An arc.plot represents enrichment bargraphs, showing abundance of positive and negative
values (e.g. log2 fold change of gene expression comparisons). Units (e.g. gene names) are
clustered and intersections are indicated by an arc.
x$up <- lapply(Map(setNames, x$log2FoldChange, x$GenesSignificant), function (i) { i[i>0] }) x$dwn <- lapply(Map(setNames, x$log2FoldChange, x$GenesSignificant), function (i) { i[i<0] }) arc.plot(x = setNames(x$LogEnrich, x$Term), up.list = x$up, down.list = x$dwn, x.mar = c(.9, .5))
Go Histogram
A bargraph showing positive and negative value abundances per category.
Supply a list of data.frames and compare each table, performing a one
sided t-test on the absolute logfc.term values.
Data is prepared as follows:
y <- lapply(leukemiasGO, function (x) { x$Upregulated <- sapply(x$log2FoldChange, function (z) sum(z>0)) x$Downregulated <- sapply(x$log2FoldChange, function (z) sum(z<0)) x }) yterms <- unique(unlist(lapply(y, function(x){ x <- subset(x, pvalCutOff <= 0.05) x <- x[order(x$LogEnrich),] head(x, 9)$GO.ID })))
An the plot generated:
par(mar = c(0,.5,2.5,8)) go.histogram(y, go.alpha.term = "pvalCutOff", gene.alpha.term = "padj", min.genes = 5, max.genes = 1e10, go.selection = yterms, show.ttest = T, main = "GO enrichment\nin leukemia differential gene expression\ncompared to control samples", axis.cex = 1, lab.cex = 1.5, main.cex = 1.5)
Visual parameters
Both functions provide a range of parameters to tweak the visual appearance of the plots.
- Font sizes of the various elements can be adjusted using the
cexparameters. - Internal margins are adjusted through
x.marandy.mar. Used in particular to accomodate the label text to the left-hand side. - Absolute scaling can be achieved through the
bar.scaleparameter, which acts as a multiplier for a predetermined average bar height. This is meant to be used in conjunction with a graphics device of constant size, to ensure visual consistency among multiple plots that vary in the number of terms displayed. - Data ranges for all output may be fixed:
elem.boundsRequire the cardinality of a term or significant subset to be in a specific range, e.g.c(10,100)-- between 10 and 100 genes.x.boundUpper limit of the value displayed on the x-axis. Forsym.plotthis is must be a value between 0 and 100. The lower limit is always 0.cell.bounds(onlycell.plot) Sets the colour scale for cell data to a fixed range. If the plot contains values outside of that range, an indicator is added to the colour legend. This is particularly useful in the event of a few extreme outliers.mid.bounds(onlysym.plot) Similar tocell.bounds, this sets the colour range of the central column cells.sym(onlycell.plot), when set toTRUE, makes the scale of cell values symmetrical.
- Colour schemes may be controlled through character vectors provided to the following parameters. Internally, the R native function
ColorRampPalette()is used.cell.col(onlycell.plot) Specify three valid colour names for cell data visualisation (lower extreme, mid point, upper extreme).mid.col(onlysym.plot) Specify two valid colour names to define the range of the central column cell colours.
- Colour keys can be adjusted in their resolution (number of boxes) using the
key.nparameter. - Gridlines may be hidden via the
gridlinesparameter.
Workflows
Gene ontology term enrichment in differential gene expression data can be performed using the Bioconductor/topGO package.
Dataset golubGO
This section provides code and comments on the workflow to perform differential
expression of genes (DEG) analysis and the gene ontology term (GO) enrichment
of from the results applied to the golub dataset from the
Bioconductor/leuke
package.
Dataset leukemiasGO
This section provides code and comments on the workflow to perform differential
expression of genes analysis and the gene ontology term enrichment
of from the results applied to the leukemiasEset dataset from the
Bioconductor/leukemiasEset package.
GO annotation was performed with the
Bioconductor/annotate
# CRAN library(parallel) library(dplyr) library(stringr) # Bioconductor library(BiocParallel) library(DESeq2) library(topGO) library(annotate) library(leukemiasEset) data(leukemiasEset) # M <- select(hu6800.db, featureNames(leukemiasEset), c("ENSEMBL","GO"), keytype = "ENSEMBL") # M <- subset(M, ONTOLOGY == "BP", c("ENSEMBL","GO")) # M <- M[!duplicated(M),] # M <- dlply(M, "ENSEMBL", function (x) unique(x$GO)) # M <- as.character(unique(unlist(select(org.Hs.eg.db,featureNames(leukemiasEset),"ENSEMBL",keytype = "ENSEMBL")))) A <- as(leukemiasEset,"data.frame") A <- subset(A, LeukemiaType %in% c("NoL","ALL", "AML", "CLL")) As <- subset(A, select = "LeukemiaType") As$LeukemiaType <- relevel(factor(as.character(As$LeukemiaType)), "NoL") A <- subset(A, select = grepl("^ENS", colnames(A))) A <- sapply(A, as.integer) A <- t(A) DEG <- DESeqDataSetFromMatrix(countData = A, colData = As, design = ~ LeukemiaType) DEG <- DESeq(DEG, fitType = "mean", parallel = T, BPPARAM = MulticoreParam(20)) n <- levels(As$LeukemiaType)[-1] leukemiasGO <- lapply(setNames(n, n), function (n) { x <- results(DEG, c("LeukemiaType", n, "NoL"), "LeukemiaType", alpha = .05) x <- as.data.frame(x) x$padj[is.na(x$padj)] <- 1 g <- new("topGOdata", ontology = "BP", description = 'Leukemia', allGenes = setNames(x$padj, rownames(x)), mapping = "org.Hs.eg.db", geneSelectionFun = function (allScore) { allScore <= 0.05 }, annotationFun = annFUN.org, ID = "Ensembl") t <- new("elimCount", testStatistic = GOFisherTest, name = "Fisher test") # test definition s <- getSigGroups(g, t) # run F-test r <- GenTable(g, pvalCutOff = s, topNodes = length(g@graph@nodes)) # return data.frame r$pvalCutOff <- as.numeric(str_replace_all(r$pvalCutOff, "[^0-9e\\-\\.]*", "")) r$LogEnrich <- log2(r$Significant / r$Expected) ga <- genesInTerm(g) # GenesAnnotated | list of genes per go-terms ga <- ga[r$GO.ID] # eliminate missing terms names(ga) <- NULL r$GenesAnnotated <- ga xs <- x[,c("padj", "log2FoldChange")] # significant stats subset xs <- subset(xs, padj < 0.05) r$GenesSignificant <- lapply(r$GenesAnnotated, intersect, rownames(xs)) # extract genes ei.rows <- mclapply(r$GenesSignificant, function (y) { if (length(y)) as.list(xs[y,,drop=FALSE]) else as.list(rep(NA_real_, length(xs))) }, mc.cores = 10) ei <- mclapply(names(xs), function(z) { lapply(ei.rows, "[[", z) }, mc.cores = 10) ei <- structure(ei, names = names(xs), row.names = seq(nrow(r)), class = "data.frame") row.names(ei) <- NULL r <- data.frame(r, ei, stringsAsFactors = FALSE, check.names = FALSE) return(r) }) # lapply(GO, function(x) {list(all(r$Annotated == sapply(r$GenesAnnotated, length)), # all(r$Significant == sapply(r$GenesSignificant, length)))}) leukemiasGO <- lapply(leukemiasGO, function(x) subset(x, pvalCutOff < 0.1)) save(leukemiasGO, file = "data/leukemiasGO.rdata")
Installation
CellPlot
Install the latest version from the repository on
github.com/dieterich-lab/CellPlot.
With the devtools package, it is an easy task:
library(devtools)
install_github('dieterich-lab/CellPlot', build_vignettes = TRUE)
Dependencies
To install the packages from Bioconducter, use the following steps:
# load bioconductor functions source("https://bioconductor.org/biocLite.R") # install packages biocLite("BiocParallel") biocLite("annotate") biocLite("DESeq2") biocLite("topGO") biocLite("leukemiasEset")
Dependencies from CRAN can be installed via:
# install packages install.packages(c("stringr","dplyr","knitr","rmarkdown","devtools"))
Session info
# session info from build machine sessionInfo()
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.