R/metaseqr.json.R
In pmoulos/metaseqR: An R package for the analysis and result reporting of RNA-Seq data by combining multiple statistical algorithms.
Defines functions
getColors
getColorScheme
getGroupColorScheme
unquote_js_fun
volcanoToJSON
filteredToJSON
biasPlotToJSON
boxplotToJSON
readNoiseToJSON
bioSaturationToJSON
bioDetectionToJSON
countsBioToJSON
mdsToJSON
#' MDS plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.mds}}.
#'
#' @param obj A list holding MDS plot data. See \code{\link{diaplot.mds}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
mdsToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
x <- obj$x
y <- obj$y
xlim <- obj$xlim
ylim <- obj$ylim
samples <- obj$samples
cols <- getColorScheme()
# Construct series
counter <- 0
series <- vector("list",length(samples))
names(series) <- names(samples)
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
series[[n]]$type="scatter"
series[[n]]$color=cols$fill[counter]
series[[n]]$marker <- list(
lineWidth=1,
states=list(
hover=list(
enabled=TRUE,
lineColor=cols$border[counter]
),
select=list(
fillColor=cols$selected[counter],
lineColor=cols$border[counter],
lineWidth=2
)
)
)
m <- match(samples[[n]],names(x))
if (length(m)>0) {
series[[n]]$data <- make.highcharts.points(x[m],y[m])
}
}
switch(jl,
highcharts = {
point.format=paste("<strong>Sample name: </strong>{point.name}<br>",
"<strong>Principal coordinate 1: </strong>{point.x}<br>",
"<strong>Principal coordinate 2: </strong>{point.y}",sep="")
json <- toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Multidimensional Scaling")
),
xAxis=list(
title=list(
useHTML=TRUE,
text="1<sup>st</sup> Principal Coordinate",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(xlim[1],3),
max=round(xlim[2],3)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="2<sup>nd</sup> Principal Coordinate",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(ylim[1],3),
max=round(ylim[2],3)
),
plotOptions=list(
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=point.format
)
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Biodetection counts plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param by Can be \code{"sample"} to create biotypes boxplots per sample or
#' \code{"biotype"} to create samples boxplots per biotype.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
countsBioToJSON <- function(obj,by=c("sample","biotype"),jl=c("highcharts")) {
by <- tolower(by[1])
jl <- tolower(jl[1])
samples <- obj$samples
status <- obj$status
altnames <- obj$altnames
counts <- round(2^obj$user$counts - 1)
counts[counts==0] <- 0.001
#counts <- obj$user$counts
covars <- obj$user$covars
biotypes <- unique(as.character(covars$biotype))
if (!is.null(altnames))
names(altnames) <- rownames(counts)
grouped <- FALSE
if (is.null(samples)) {
if (is.null(colnames(counts)))
samplenames <- paste("Sample",1:ncol(counts),sep=" ")
else
samplenames <- colnames(counts)
samples <- list(Samples=nams)
}
else if (is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
grouped <- TRUE
}
# y label formatter for logarithmic axis
y.label.formatter <- paste('function() {if(this.value === 0.001)',
'{return 0;} else {return Highcharts.Axis.prototype.',
'defaultLabelFormatter.call(this);}}',sep="")
tooltip.point.formatter <- paste("function() {",
" var min = this.low === 0.001 ? 0 : this.low;" ,
" var q1 = this.q1 === 0.001 ? 0 : this.q1;" ,
" var med = this.median === 0.001 ? 0 : this.median;",
" var q3 = this.q3 === 0.001 ? 0 : this.q3;",
" var max = this.high === 0.001 ? 0 : this.high;",
" var str = 'Maximum: ' + max + '<br/>' +",
" 'Upper quartile: ' + q3 + '<br/>' +",
" 'Median: ' + med + '<br/>' +",
" 'Lower quartile: ' + q1 + '<br/>' +",
" 'Minimum: ' + min + '<br/>';",
" return str;",
"}",sep="")
# Legend clicker
boxplot.onclick <- paste("function() { ",
" var chart = this.chart;",
" var outlier_id = chart.get(this.name);",
" if (!outlier_id.visible) {",
" outlier_id.show();",
" } else {",
" outlier_id.hide();",
" }",
"}",sep="")
# Outliers tooltip
if (is.null(obj$altnames)) {
outlier.pointformat <- paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
else {
outlier.pointformat <- paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Gene name: {point.alt_name}<br/>',
'Value: {point.y}<br/>',
sep=""
)
}
if (by=="sample") {
cols <- getColorScheme(length(biotypes))
box.list <- json <- vector("list",length(samplenames))
names(box.list) <- names(json) <- samplenames
for (n in samplenames) {
box.list[[n]] <- vector("list",length(biotypes))
names(box.list[[n]]) <- biotypes
for (b in biotypes)
box.list[[n]][[b]] <- counts[covars$biotype==b,n]
B <- boxplot(box.list[[n]],plot=FALSE)$stats
colnames(B) <- biotypes
o.list <- lapply(names(box.list[[n]]),function(x,M,b) {
v <- b[,x]
o <- which(M[[x]]<v[1] | M[[x]]>v[5])
if (length(o)>0)
return(M[[x]][o])
else
return(NULL)
},box.list[[n]],B)
names(o.list) <- biotypes
# Data series
BB <- matrix(0,nrow(B),ncol(B)) # Workaround of strange problem...
colnames(BB) <- colnames(B)
for (jj in 1:ncol(B))
BB[,jj] <- round(B[,jj],3)
d <- as.data.frame(BB)
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(biotypes))
names(series) <- biotypes
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$name <- s
series[[s]]$color <- cols$fill[counter]
#series[[s]]$turboThreshold <- 10000
series[[s]]$data <- list(unname(as.list(d[,s])))
r <- round(d[,s])
series[[s]]$tooltip=list(
pointFormat=paste('<strong>Population: ',
length(box.list[[n]][[s]]),'</strong><br/>',
'Maximum: ',r[6],'<br/>',
'Upper quartile: ',r[5],'<br/>',
'Median: ',r[4],'<br/>',
'Lower quartile: ',r[3],'<br/>',
'Minimum: ',r[2],'<br/>',sep="")
)
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(biotypes))
names(outliers) <- biotypes
for (o in names(outliers)) {
counter <- counter + 1
outliers[[o]] <- list()
outliers[[o]]$id <- o
outliers[[o]]$name <- o
outliers[[o]]$type <- "scatter"
outliers[[o]]$showInLegend <- FALSE
#outliers[[o]]$turboThreshold <- 10000
outliers[[o]]$color <- cols$fill[counter]
outliers[[o]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
outliers[[o]]$data <- list()
x <- rep(d[1,o],length(o.list[[o]]))
names(x) <- names(o.list[[o]])
if (is.null(obj$altnames)) {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[o]])
}
else {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[o]],
unname(altnames[names(x)]))
}
}
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Biotype detection for sample ",n,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
categories=biotypes,
title=list(
text="Biotype",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
type="logarithmic",
showFirstLabel=FALSE,
min=1e-4,
tickInterval=1,
title=list(
useHTML=TRUE,
#text="Read count (log<sub>2</sub>)",
text="Expression (read count)",
margin=25,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
),
formatter=y.label.formatter
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
)
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
}
return(unquote_js_fun(json))
}
else if (by=="biotype") {
cols <- getColorScheme(length(samples))
box.list <- json <- vector("list",length(biotypes))
names(box.list) <- names(json) <- biotypes
for (b in biotypes) {
box.list[[b]] <- vector("list",length(samplenames))
names(box.list[[b]]) <- samplenames
for (n in samplenames)
box.list[[b]][[n]] <- counts[covars$biotype==b,n]
B <- boxplot(box.list[[b]],plot=FALSE)$stats
colnames(B) <- samplenames
o.list <- lapply(names(box.list[[b]]),function(x,M,b) {
v <- b[,x]
o <- which(M[[x]]<v[1] | M[[x]]>v[5])
if (length(o)>0)
return(M[[x]][o])
else
return(NULL)
},box.list[[b]],B)
names(o.list) <- samplenames
# Data series
BB <- matrix(0,nrow(B),ncol(B)) # Workaround of strange problem...
colnames(BB) <- colnames(B)
for (jj in 1:ncol(B))
BB[,jj] <- round(B[,jj],3)
d <- as.data.frame(BB)
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(samples))
names(series) <- names(samples)
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$name=s
if (grouped)
series[[s]]$color=cols$fill[counter]
else
series[[s]]$color=cols$fill[1]
m <- match(samples[[s]],colnames(d))
series[[s]]$data <- unname(as.list(d[,m]))
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(samples))
names(outliers) <- names(samples)
for (o in names(outliers)) {
counter <- counter + 1
outliers[[o]] <- list()
outliers[[o]]$id <- o
outliers[[o]]$name <- o
outliers[[o]]$type <- "scatter"
outliers[[o]]$showInLegend <- FALSE
if (grouped) {
outliers[[o]]$color <- cols$fill[counter]
outliers[[o]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
}
else {
outliers[[o]]$color <- cols$fill[1]
outliers[[o]]$marker <- list(
fillColor=cols$fill[1],
symbol="circle",
lineWidth=1,
lineColor=cols$border[1]
)
}
outliers[[o]]$data <- list()
m <- match(samples[[o]],colnames(d))
if (length(m)>0) {
for (i in m) {
x <- rep(d[1,i],length(o.list[[i]]))
names(x) <- names(o.list[[i]])
if (is.null(obj$altnames)) {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[i]])
}
else {
outliers[[o]]$data <- c(outliers[[o]]$data,
make.highcharts.points(x,o.list[[i]],
unname(altnames)))
}
}
}
}
json[[b]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Detection for biotype ",b,
" (population: ",lengths(box.list[[b]])[1],
")",sep="")
),
legend=list(
enabled=TRUE
),
xAxis=list(
categories=samplenames,
title=list(
text="Sample name",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
type="logarithmic",
showFirstLabel=FALSE,
min=1e-4,
tickInterval=1,
title=list(
text="Expression (read count)",
margin=25,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
),
formatter=y.label.formatter
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Condition {series.name}</span>',
'<br/>',
'<span style="font-weight:bold;">',
'Sample {point.key}',
'</span><br/>',sep=""
),
pointFormatter=tooltip.point.formatter
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Condition {series.name}</span>',
'<br/>',sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
}
return(unquote_js_fun(json))
}
}
#' Biodetection barplot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
bioDetectionToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
samples <- obj$samples
status <- obj$status
plotdata <- obj$user$plotdata
covars <- obj$user$covars
if (!is.null(samples)&& is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
names(plotdata$biotables) <- samplenames
}
# Otherwise we are using the names present in the input object
abu <- which(plotdata$genome>7)
nabu <- which(plotdata$genome<=7)
cols <- getColorScheme()
json <- vector("list",length(samplenames))
names(json) <- samplenames
for (n in samplenames) {
# Data series
series.abu <- vector("list",3)
names(series.abu) <- c("genome","detectionVSgenome","detectionVSsample")
series.abu$genome <- list()
series.abu$genome$id <- "abu_genome"
series.abu$genome$name <- "% in genome"
series.abu$genome$color <- cols$trans[1]
series.abu$genome$pointPlacement <- -0.2
series.abu$genome$data <- round(as.numeric(plotdata$genome[abu]),3)
series.abu$detectionVSgenome <- list()
series.abu$detectionVSgenome$id <- "abu_detected"
series.abu$detectionVSgenome$name <- "% detected"
series.abu$detectionVSgenome$color <- cols$trans[2]
series.abu$detectionVSgenome$pointPlacement <- 0
series.abu$detectionVSgenome$data <- round(as.numeric(
plotdata$biotables[[n]][1,abu]),3)
series.abu$detectionVSsample <- list()
series.abu$detectionVSsample$id <- "abu_sample"
series.abu$detectionVSsample$name <- "% in sample"
series.abu$detectionVSsample$color <- cols$trans[3]
series.abu$detectionVSsample$pointPlacement <- 0.2
series.abu$detectionVSsample$data <- round(as.numeric(
plotdata$biotables[[n]][2,abu]),3)
series.nabu <- vector("list",3)
names(series.nabu) <- c("genome","detectionVSgenome",
"detectionVSsample")
series.nabu$genome <- list()
series.nabu$genome$name <- "% in genome"
series.nabu$genome$yAxis <- 1
series.nabu$genome$pointStart <- length(abu)
series.nabu$genome$linkedTo <- "abu_genome"
series.nabu$genome$color <- cols$trans[1]
series.nabu$genome$pointPlacement <- -0.2
series.nabu$genome$data <- round(as.numeric(plotdata$genome[nabu]),3)
series.nabu$detectionVSgenome <- list()
series.nabu$detectionVSgenome$name <- "% detected"
series.nabu$detectionVSgenome$yAxis <- 1
series.nabu$detectionVSgenome$pointStart <- length(abu)
series.nabu$detectionVSgenome$linkedTo <- "abu_detected"
series.nabu$detectionVSgenome$color <- cols$trans[2]
series.nabu$detectionVSgenome$pointPlacement <- 0
series.nabu$detectionVSgenome$data <- round(as.numeric(
plotdata$biotables[[n]][1,nabu]),3)
series.nabu$detectionVSsample <- list()
series.nabu$detectionVSsample$name <- "% in sample"
series.nabu$detectionVSsample$yAxis <- 1
series.nabu$detectionVSsample$pointStart <- length(abu)
series.nabu$detectionVSsample$linkedTo <- "abu_sample"
series.nabu$detectionVSsample$color <- cols$trans[3]
series.nabu$detectionVSsample$pointPlacement <- 0.2
series.nabu$detectionVSsample$data <- round(as.numeric(
plotdata$biotables[[n]][2,nabu]),3)
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="column",
alignTicks=FALSE
),
title=list(
text=paste("Comparative biotype detection for ",
"sample ",n,sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
tooltip=list(
shared=TRUE
),
xAxis=list(
categories=names(plotdata$genome)[c(abu,nabu)],
title=list(
text="Biotype",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
),
plotLines=list(
list(
color="#8A8A8A",
width=1.5,
dashStyle="Dash",
value=length(abu)-0.5
)
),
plotBands=list(
list(
color="#FFFFE0",
from=-0.5,
to=length(abu)-0.5
),
list(
color="#FFECEB",
from=length(abu)-0.5,
to=length(plotdata$genome)
)
)
),
yAxis=list(
list(
min=0,
max=70,
title=list(
text="% of abundant features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
list(
min=0,
max=7,
title=list(
text="% of non-abundant features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
opposite=TRUE
)
),
plotOptions=list(
column=list(
grouping=FALSE,
shadow=FALSE,
groupPadding=0.3,
pointPadding=0.25,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'font-weight:bold">',
'{point.key}</span><br/>',sep=""
)
)
)
),
series=c(unname(series.abu),unname(series.nabu))
)
)
}
)
}
return(json)
}
#' Biotype saturation plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param by Can be \code{"sample"} to create biotypes boxplots per sample or
#' \code{"biotype"} to create samples boxplots per biotype.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
bioSaturationToJSON <- function(obj,by=c("sample","biotype"),
jl=c("highcharts")) {
by <- tolower(by[1])
jl <- tolower(jl[1])
samples <- obj$samples
plotdata <- obj$user$plotdata
if (!is.null(samples)&& is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
names(plotdata) <- samplenames
}
if (by=="sample") {
json <- vector("list",length(samplenames))
names(json) <- samplenames
for (n in samplenames) {
depth <- round(plotdata[[n]][,1]/1e+6)
global <- round(plotdata[[n]][,2])
M <- plotdata[[n]][,3:ncol(plotdata[[n]])]
# To determine the separation
ord <- sort(M[nrow(M),],decreasing=TRUE,index.return=TRUE)
abu <- ord$ix[1:2]
names(abu) <- names(ord$x[1:2])
nabu <- ord$ix[3:length(ord$ix)]
names(nabu) <- names(ord$x[3:length(ord$x)])
cols <- getColorScheme(ncol(plotdata[[n]])-1)
counter <- 1
global.series <- list(
global=list(
id="global",
name="global",
color=cols$fill[counter],
data=make.highcharts.points(depth,global)
)
)
abu.series <- vector("list",2)
names(abu.series) <- names(abu)
for (s in names(abu.series)) {
counter <- counter + 1
abu.series[[s]] <- list()
abu.series[[s]]$id <- s
abu.series[[s]]$name <- s
abu.series[[s]]$color <- cols$fill[counter]
abu.series[[s]]$data <- make.highcharts.points(depth,
round(M[,s]))
}
nabu.series <- vector("list",length(3:ncol(M)))
names(nabu.series) <- names(nabu)
for (s in names(nabu.series)) {
counter <- counter + 1
nabu.series[[s]] <- list()
nabu.series[[s]]$id <- s
nabu.series[[s]]$name <- s
nabu.series[[s]]$color <- cols$fill[counter]
nabu.series[[s]]$data <- make.highcharts.points(depth,
round(M[,s]))
}
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Biotype saturations for sample ",n,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
title=list(
text="Read depth (millions of reads)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
min=0,
max=max(global),
title=list(
text="Detected features",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
series=list(
lineWidth=2
),
scatter=list(
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
),
pointFormat=paste(
"Depth: {point.x}M<br/>",
"Detected features: {point.y}",
sep="")
)
)
),
series=c(unname(global.series),
unname(abu.series),unname(nabu.series))
)
)
}
)
}
return(json)
}
else if (by=="biotype") {
biotypes <- colnames(plotdata[[1]])[2:ncol(plotdata[[1]])]
depths <- vector("list",length(plotdata))
names(depths) <- samplenames
for (n in samplenames)
depths[[n]] <- round(plotdata[[n]][,1]/1e+6)
json <- vector("list",length(biotypes))
names(json) <- biotypes
for (b in biotypes) {
series <- vector("list",length(plotdata))
names(series) <- samplenames
cols <- getColorScheme(length(samplenames))
counter <- 0
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$id <- s
series[[s]]$name <- s
series[[s]]$color <- cols$fill[counter]
series[[s]]$data <- make.highcharts.points(depths[[s]],
round(plotdata[[s]][,b]))
}
json[[b]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Sample saturations for biotype ",b,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
title=list(
text="Read depth (millions of reads)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
title=list(
text="Detected features",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
series=list(
lineWidth=2
),
scatter=list(
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Sample {series.name}</span><br/>',
sep=""
),
pointFormat=paste(
"Depth: {point.x}M<br/>",
"Detected features: {point.y}",
sep="")
)
)
),
series=c(unname(series))
)
)
}
)
}
return(json)
}
}
#' Read noise plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.noiseq}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
readNoiseToJSON <- function(obj,jl=c("highcharts"),seed=42) {
jl <- tolower(jl[1])
d <- obj$user
samples <- obj$samples
# Too many points for a lot of curves of interactive data
if (nrow(d)>1000) {
set.seed(seed)
ii <- sort(sample(1:nrow(d),998))
ii <- c(1,ii,nrow(d))
d <- cbind(d[ii,1],d[ii,2:ncol(d)])
}
if (is.null(samples))
samples <- 1:(ncol(d)-1)
if (is.numeric(samples))
samplenames = colnames(dat)[samples+1]
if (is.list(samples))
samplenames <- unlist(samples)
cols <- getColorScheme(length(samplenames))
# Construct series
counter <- 0
series <- vector("list",length(samplenames))
names(series) <- samplenames
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
series[[n]]$color=cols$fill[counter]
series[[n]]$data <- make.highcharts.points(d[,1],d[,n])
series[[n]]$tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
)
}
switch(jl,
highcharts = {
json <- toJSON(list(
chart=list(
type="line",
zoomType="xy"
),
title=list(
text=paste("RNA-Seq mapped reads noise")
),
xAxis=list(
title=list(
text="% detected features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=0,
max=100
),
yAxis=list(
title=list(
text="% of total reads",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
tickPositions=seq(0,110,10)
),
plotOptions=list(
line=list(
allowPointSelect=TRUE,
lineWidth=1,
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
),
turboThreshold=50000
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Boxplots JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.boxplot}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.boxplot}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
boxplotToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
b <- obj$plot
name <- obj$samples
status <- obj$status
altnames <- obj$altnames
o.list <- obj$user
grouped <- FALSE
if (is.null(name)) {
if (is.null(colnames(b$stats)))
nams <- paste("Sample",1:ncol(b$stats),sep=" ")
else
nams <- colnames(b$stats)
name <- list(Samples=nams)
}
else if (length(name)==1 && name=="none") {
nams <- rep("",ncol(b$stats))
name <- list(Samples=nams)
}
else if (is.list(name)) { # Is sample.list
nams <- unlist(name,use.names=FALSE)
grouped <- TRUE
}
cols <- getColorScheme()
# Data series
d <- as.data.frame(round(b$stat,3))
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
colnames(d) <- nams
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(name))
names(series) <- names(name)
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
if (grouped)
series[[n]]$color=cols$fill[counter]
else
series[[n]]$color=cols$fill[1]
m <- match(name[[n]],colnames(d))
series[[n]]$data <- unname(as.list(d[,m]))
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(name))
names(outliers) <- names(name)
for (n in names(outliers)) {
counter <- counter + 1
outliers[[n]] <- list()
outliers[[n]]$id <- n
outliers[[n]]$name <- n
outliers[[n]]$type <- "scatter"
outliers[[n]]$showInLegend <- FALSE
if (grouped) {
outliers[[n]]$color <- cols$fill[counter]
outliers[[n]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
}
else {
outliers[[n]]$color <- cols$fill[1]
outliers[[n]]$marker <- list(
fillColor=cols$fill[1],
symbol="circle",
lineWidth=1,
lineColor=cols$border[1]
)
}
outliers[[n]]$data <- list()
m <- match(name[[n]],colnames(d))
if (length(m)>0) {
for (i in m)
x <- rep(d[1,i],length(o.list[[i]]))
names(x) <- names(o.list[[i]])
outliers[[n]]$data <- c(outliers[[n]]$data,
make.highcharts.points(x,o.list[[i]],unname(altnames)))
}
}
# Boxplot tooltip point formatter for the case of zeros
tooltip.point.formatter <- paste("function() {",
" var min = this.low === 0.001 ? 0 : this.low;" ,
" var q1 = this.q1 === 0.001 ? 0 : this.q1;" ,
" var med = this.median === 0.001 ? 0 : this.median;",
" var q3 = this.q3 === 0.001 ? 0 : this.q3;",
" var max = this.high === 0.001 ? 0 : this.high;",
" var str = 'Maximum: ' + max + '<br/>' +",
" 'Upper quartile: ' + q3 + '<br/>' +",
" 'Median: ' + med + '<br/>' +",
" 'Lower quartile: ' + q1 + '<br/>' +",
" 'Minimum: ' + min + '<br/>';",
" return str;",
"}",sep="")
# Legend clicker
boxplot.onclick <- paste("function() {",
" var chart = this.chart;",
" var outlier_id = chart.get(this.name);",
" if (!outlier_id.visible) {",
" outlier_id.show();",
" } else {",
" outlier_id.hide();",
" }",
"}",sep="")
if (is.null(obj$altnames)) {
outlier.pointformat=paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
else {
outlier.pointformat=paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Gene name: {point.alt_name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
json <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Boxplot ",status,sep="")
),
legend=list(
enabled=TRUE
),
xAxis=list(
categories=nams,
title=list(
text="Sample name",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Read count (log<sub>2</sub>)",
margin=30,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Condition {series.name}</span><br/>',
'<span style="font-weight:bold">',
'Sample {point.key}</span><br/>',sep=""
),
pointFormatter=tooltip.point.formatter
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Condition {series.name}</span><br/>',
sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
return(unquote_js_fun(json))
}
#' GC/length bias plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.edaseq}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
biasPlotToJSON <- function(obj,jl=c("highcharts"),seed=1) {
jl <- tolower(jl[1])
counts <- round(nat2log(obj$user$counts),3)
status <- obj$status
covar <- obj$user$covar
covarname <- obj$user$covarname
samples <- obj$samples
# Too many points for a lot of curves of interactive data
if (nrow(counts)>2000) {
set.seed(seed)
ii <- sample(1:nrow(counts),2000)
counts <- counts[ii,]
covar <- covar[ii]
}
# If length bias, not nice to have x-axis at -200k
min.x <- ifelse(max(covar>100),0,"undefined")
if (is.null(samples)) {
if (is.null(colnames(x)))
samplenames <- paste("Sample",1:ncol(counts),sep=" ")
else
samplenames <- colnames(counts)
samples <- list(Samples=nams)
cols <- getColorScheme(length(samples))
}
else if (is.list(samples)) { # Is sample.list
samplenames <- unlist(samples,use.names=FALSE)
grouped <- TRUE
}
colnames(counts) <- samplenames
# Construct series
counter <- 0
series <- vector("list",length(samplenames))
names(series) <- samplenames
for (n in names(series)) {
counter <- counter + 1
x <- counts[,n]
fit <- lowess(covar,x)
series[[n]] <- list()
series[[n]]$name <- n
series[[n]]$color <- cols$fill[counter]
series[[n]]$data <- lapply(1:length(x),function(i,x,y) {
return(c(x[i],y[i])) },round(fit$x,3),round(fit$y,3))
}
switch(jl,
highcharts = {
json <- toJSON(list(
chart=list(
type="line",
zoomType="xy"
),
title=list(
text=paste(covarname," bias detection - ",status)
),
xAxis=list(
min=min.x,
title=list(
text=covarname,
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Read count (log<sub>2</sub>)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE
),
plotOptions=list(
line=list(
marker=list(
enabled=FALSE,
states=list(
hover=list(
enabled=FALSE
)
)
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
),
turboThreshold=50000
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Filtered genes barplot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.filtered}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.filered}}.
#' @param by Either \code{"chromosome"} or \code{"biotype"}
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
filteredToJSON <- function(obj,by=c("chromosome","biotype"),
jl=c("highcharts")) {
jl <- tolower(jl[1])
by <- tolower(by[1])
filtered <- obj$user$filtered
total <- obj$user$total
cols <- getColorScheme(2)
if (by=="chromosome") {
chr <- table(as.character(filtered$chromosome))
chr.all <- table(as.character(total$chromosome))
barlab.chr <- as.character(chr)
per.chr <- round(chr/chr.all[names(chr)],3)
per.chr[per.chr>1] <- 1
series <- vector("list",2)
names(series) <- c("number","fraction")
series$number <- list()
series$number$id <- "chr_number"
series$number$name <- "Number of genes"
series$number$color <- cols$fill[1]
series$number$pointPlacement <- -0.2
series$number$data <- unname(chr)
series$fraction <- list()
series$fraction$id <- "chr_fraction"
series$fraction$name <- "Fraction of total genes"
series$fraction$color <- cols$fill[2]
series$fraction$pointPlacement <- 0.2
series$fraction$yAxis <- 1
series$fraction$data <- unname(per.chr)
what <- chr
}
else if (by=="biotype") {
bt <- table(as.character(filtered$biotype))
bt.all <- table(as.character(total$biotype))
barlab.bt <- as.character(bt)
per.bt <- round(bt/bt.all[names(bt)],3)
per.bt[per.bt>1] <- 1
series <- vector("list",2)
names(series) <- c("number","fraction")
series$number <- list()
series$number$id <- "bt_number"
series$number$name <- "Number of genes"
series$number$color <- cols$fill[1]
series$number$pointPlacement <- -0.2
series$number$data <- unname(bt)
series$fraction <- list()
series$fraction$id <- "bt_fraction"
series$fraction$name <- "Fraction of total genes"
series$fraction$color <- cols$fill[2]
series$fraction$pointPlacement <- 0.2
series$fraction$yAxis <- 1
series$fraction$data <- unname(per.bt)
what <- bt
}
json <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="column",
alignTicks=FALSE
),
title=list(
text=paste("Filtered genes per ",by,sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
tooltip=list(
shared=TRUE
),
xAxis=list(
categories=names(what),
title=list(
text=paste(toupper(substr(by,1,1)),substr(by,2,
nchar(by)),sep=""),
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
list(
lineColor=cols$fill[1],
lineWidth=2,
min=0,
tickAmount=11,
title=list(
text="Number of genes",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
offset=10
),
list(
lineColor=cols$fill[2],
lineWidth=2,
min=0,
max=1,
tickAmount=11,
#tickInterval=0.1,
title=list(
text="Fraction of total genes",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
opposite=TRUE,
offset=10
)
),
plotOptions=list(
column=list(
grouping=FALSE,
shadow=FALSE,
groupPadding=0.2,
pointPadding=0.2,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'font-weight:bold">',
'{point.key}</span><br/>',sep=""
)
)
)
),
series=c(unname(series))
)
)
}
)
return(json)
}
#' Volcano JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.volcano}}.
#'
#' @param obj A list holding volcano plot data. See \code{\link{diaplot.volcano}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
volcanoToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
f <- obj$x
p <- obj$y
xlim <- obj$xlim
ylim <- obj$ylim
alt.names <- obj$altnames
pcut <- obj$pcut
fcut <- obj$fcut
up <- obj$user$up
down <- obj$user$down
ff <- obj$user$unf
pp <- obj$user$unp
alt.names.neutral <- obj$user$ualt
con <- obj$user$con
switch(jl,
highcharts = {
if (is.null(alt.names))
point.format=paste("<strong>Gene ID: </strong>{point.name}<br>",
"<strong>Fold change: </strong>{point.x}<br>",
"<strong>Significance: </strong>{point.y}",sep="")
else
point.format=paste("<strong>Gene name: </strong>",
"{point.alt_name}<br>",
"<strong>Gene ID: </strong>{point.name}<br>",
"<strong>Fold change: </strong>{point.x}<br>",
"<strong>Significance: </strong>{point.y}",sep="")
json <- toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Volcano plot for",con)
),
xAxis=list(
title=list(
text="Fold change",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(xlim[1],3),
max=round(xlim[2],3)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Significance (-log<sub>10</sub>(p-value))",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(ylim[1]-2,3),
max=round(ylim[2],3)
),
#legend=list(
# layout="vertical",
# align="left",
# verticalAlign="top",
# floating=TRUE,
# backgroundColor="#FFFFFF",
# borderWidth=1
#),
plotOptions=list(
scatter=list(
allowPointSelect=TRUE,
marker=list(
radius=2,
states=list(
hover=list(
enabled=TRUE,
lineColor="#333333"
)
)
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=point.format
),
turboThreshold=50000
)
),
series=list(
list(
name="Up-regulated",
color="#EE0000",
marker=list(
symbol="circle"
),
data=make.highcharts.points(f[up],-log10(p[up]),
unname(alt.names[up]))
),
list(
name="Down-regulated",
marker=list(
symbol="circle"
),
color="#00CD00",
data=make.highcharts.points(f[down],-log10(p[down]),
unname(alt.names[down]))
),
list(
name="Unregulated",
marker=list(
symbol="circle"
),
color="#0000EE",
data=make.highcharts.points(ff,-log10(pp),
unname(alt.names.neutral))
),
list(
name="Downfold threshold",
color="#000000",
type="line",
dashStyle="dash",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.x}<br/>',sep="")
),
data=list(round(c(-fcut,ylim[1]-5),3),
round(c(-fcut,ylim[2]),3))
),
list(
name="Upfold threshold",
color="#000000",
type="line",
dashStyle="Dash",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.x}<br/>',sep="")
),
data=list(round(c(fcut,ylim[1]-5),3),
round(c(fcut,ylim[2]),3))
),
list(
name="Significance threshold",
color="#000000",
type="line",
dashStyle="DashDot",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.y}<br/>',sep="")
),
data=list(round(c(xlim[1],-log10(pcut)),3),
round(c(xlim[2],-log10(pcut)),3))
)
)
)
)
}
)
return(json)
}
unquote_js_fun <- function(js) {
if (is.list(js))
js <- lapply(js,unquote_js_fun)
else {
op <- gregexpr(pattern="function",js)
cl <- gregexpr(pattern="}\\\"",js)
if (length(op)>0) {
starts <- as.numeric(op[[1]])
for (i in 1:length(starts))
substr(js,starts[i]-1,starts[i]-1) <- " "
ends <- as.numeric(cl[[1]])
for (i in 1:length(starts))
substr(js,ends[i]+1,ends[i]+1) <- " "
}
}
return(js)
}
getGroupColorScheme <- function(group) {
cols <- getColorScheme(length(group))
classes <- as.factor(as.class.vector(group))
design <- as.numeric(classes)
return(lapply(cols,function(x,classes,design) {
return(x[1:length(levels(classes))][design])
},classes,design))
}
getColorScheme <- function(n=NULL) {
if (missing(n) || is.null(n))
return(getColors())
else {
cols <- getColors()
if (n > length(cols$fill)) {
cols$fill <- rep(cols$fill,length.out=n)
cols$border <- rep(cols$border,length.out=n)
cols$select <- rep(cols$select,length.out=n)
cols$trans <- rep(cols$trans,length.out=n)
}
return(cols)
}
}
getColors <- function() {
return(list(
fill=c("#CD0000","#00CD00","#0000EE","#FFD700","#87CEEB","#CD8500",
"#DEB887","#FF0000","#0000FF","#00FF00","#FFA500","#A9A9A9",
"#008B00","#313131","#FFC0CB","#A52A2A","#FF00FF","#9ACD32",
"#8B636C","#2E8B57","#008B8B"),
border=c("#850000","#006B00","#000085","#927C00","#156280","#5A3A00",
"#8B7457","#935E18","#000080","#008500","#603E00","#454545",
"#073E07","#000000","#896067","#691111","#7C007C","#3A4D14",
"#5B1726","#0C2517","#062A2A"),
selected=c("#FF0000","#00FF00","#0066FF","#FFD700","#FFEB77","#FFB428",
"#FFD9A5","#FF326D","#0089FF","#B3FF00","#FFC352","#D9D9D9",
"#00EC00","#8E8E8E","#FFDAE0","#F94444","#FF87FF","#C2FF45",
"#EA889D","#4EE590","#00DADA"),
trans=c("rgba(205,0,0,0.6)","rgba(0,205,0,0.6)","rgba(0,0,238,0.6)",
"rgba(255,215,0,0.6)","rgba(135,206,235,0.6)","rgba(205,133,0,0.6)",
"rgba(222,184,135,0.6)","rbga(255,0,0,0.5)","rgba(0,0,255,0.5)",
"rgba(0,255,0,0.5)","rgba(255,165,0,0.6)","rgba(169,169,169,0.5)",
"rgba(0,139,0,0.6)","rgba(49,49,49,0.6)","rgba(255,192,203,0.5)",
"rgba(165,42,42,0.6)","rgba(255,0,255,0.6)","rgba(154,205,50,0.6)",
"rgba(139,99,108,0.6)","rgba(46,139,87,0.6)","rgba(0,139,139,0.6)")
))
}
pmoulos/metaseqR documentation built on Dec. 21, 2020, 6:19 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.
Defines functions getColors getColorScheme getGroupColorScheme unquote_js_fun volcanoToJSON filteredToJSON biasPlotToJSON boxplotToJSON readNoiseToJSON bioSaturationToJSON bioDetectionToJSON countsBioToJSON mdsToJSON
#' MDS plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.mds}}.
#'
#' @param obj A list holding MDS plot data. See \code{\link{diaplot.mds}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
mdsToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
x <- obj$x
y <- obj$y
xlim <- obj$xlim
ylim <- obj$ylim
samples <- obj$samples
cols <- getColorScheme()
# Construct series
counter <- 0
series <- vector("list",length(samples))
names(series) <- names(samples)
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
series[[n]]$type="scatter"
series[[n]]$color=cols$fill[counter]
series[[n]]$marker <- list(
lineWidth=1,
states=list(
hover=list(
enabled=TRUE,
lineColor=cols$border[counter]
),
select=list(
fillColor=cols$selected[counter],
lineColor=cols$border[counter],
lineWidth=2
)
)
)
m <- match(samples[[n]],names(x))
if (length(m)>0) {
series[[n]]$data <- make.highcharts.points(x[m],y[m])
}
}
switch(jl,
highcharts = {
point.format=paste("<strong>Sample name: </strong>{point.name}<br>",
"<strong>Principal coordinate 1: </strong>{point.x}<br>",
"<strong>Principal coordinate 2: </strong>{point.y}",sep="")
json <- toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Multidimensional Scaling")
),
xAxis=list(
title=list(
useHTML=TRUE,
text="1<sup>st</sup> Principal Coordinate",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(xlim[1],3),
max=round(xlim[2],3)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="2<sup>nd</sup> Principal Coordinate",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(ylim[1],3),
max=round(ylim[2],3)
),
plotOptions=list(
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=point.format
)
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Biodetection counts plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param by Can be \code{"sample"} to create biotypes boxplots per sample or
#' \code{"biotype"} to create samples boxplots per biotype.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
countsBioToJSON <- function(obj,by=c("sample","biotype"),jl=c("highcharts")) {
by <- tolower(by[1])
jl <- tolower(jl[1])
samples <- obj$samples
status <- obj$status
altnames <- obj$altnames
counts <- round(2^obj$user$counts - 1)
counts[counts==0] <- 0.001
#counts <- obj$user$counts
covars <- obj$user$covars
biotypes <- unique(as.character(covars$biotype))
if (!is.null(altnames))
names(altnames) <- rownames(counts)
grouped <- FALSE
if (is.null(samples)) {
if (is.null(colnames(counts)))
samplenames <- paste("Sample",1:ncol(counts),sep=" ")
else
samplenames <- colnames(counts)
samples <- list(Samples=nams)
}
else if (is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
grouped <- TRUE
}
# y label formatter for logarithmic axis
y.label.formatter <- paste('function() {if(this.value === 0.001)',
'{return 0;} else {return Highcharts.Axis.prototype.',
'defaultLabelFormatter.call(this);}}',sep="")
tooltip.point.formatter <- paste("function() {",
" var min = this.low === 0.001 ? 0 : this.low;" ,
" var q1 = this.q1 === 0.001 ? 0 : this.q1;" ,
" var med = this.median === 0.001 ? 0 : this.median;",
" var q3 = this.q3 === 0.001 ? 0 : this.q3;",
" var max = this.high === 0.001 ? 0 : this.high;",
" var str = 'Maximum: ' + max + '<br/>' +",
" 'Upper quartile: ' + q3 + '<br/>' +",
" 'Median: ' + med + '<br/>' +",
" 'Lower quartile: ' + q1 + '<br/>' +",
" 'Minimum: ' + min + '<br/>';",
" return str;",
"}",sep="")
# Legend clicker
boxplot.onclick <- paste("function() { ",
" var chart = this.chart;",
" var outlier_id = chart.get(this.name);",
" if (!outlier_id.visible) {",
" outlier_id.show();",
" } else {",
" outlier_id.hide();",
" }",
"}",sep="")
# Outliers tooltip
if (is.null(obj$altnames)) {
outlier.pointformat <- paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
else {
outlier.pointformat <- paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Gene name: {point.alt_name}<br/>',
'Value: {point.y}<br/>',
sep=""
)
}
if (by=="sample") {
cols <- getColorScheme(length(biotypes))
box.list <- json <- vector("list",length(samplenames))
names(box.list) <- names(json) <- samplenames
for (n in samplenames) {
box.list[[n]] <- vector("list",length(biotypes))
names(box.list[[n]]) <- biotypes
for (b in biotypes)
box.list[[n]][[b]] <- counts[covars$biotype==b,n]
B <- boxplot(box.list[[n]],plot=FALSE)$stats
colnames(B) <- biotypes
o.list <- lapply(names(box.list[[n]]),function(x,M,b) {
v <- b[,x]
o <- which(M[[x]]<v[1] | M[[x]]>v[5])
if (length(o)>0)
return(M[[x]][o])
else
return(NULL)
},box.list[[n]],B)
names(o.list) <- biotypes
# Data series
BB <- matrix(0,nrow(B),ncol(B)) # Workaround of strange problem...
colnames(BB) <- colnames(B)
for (jj in 1:ncol(B))
BB[,jj] <- round(B[,jj],3)
d <- as.data.frame(BB)
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(biotypes))
names(series) <- biotypes
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$name <- s
series[[s]]$color <- cols$fill[counter]
#series[[s]]$turboThreshold <- 10000
series[[s]]$data <- list(unname(as.list(d[,s])))
r <- round(d[,s])
series[[s]]$tooltip=list(
pointFormat=paste('<strong>Population: ',
length(box.list[[n]][[s]]),'</strong><br/>',
'Maximum: ',r[6],'<br/>',
'Upper quartile: ',r[5],'<br/>',
'Median: ',r[4],'<br/>',
'Lower quartile: ',r[3],'<br/>',
'Minimum: ',r[2],'<br/>',sep="")
)
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(biotypes))
names(outliers) <- biotypes
for (o in names(outliers)) {
counter <- counter + 1
outliers[[o]] <- list()
outliers[[o]]$id <- o
outliers[[o]]$name <- o
outliers[[o]]$type <- "scatter"
outliers[[o]]$showInLegend <- FALSE
#outliers[[o]]$turboThreshold <- 10000
outliers[[o]]$color <- cols$fill[counter]
outliers[[o]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
outliers[[o]]$data <- list()
x <- rep(d[1,o],length(o.list[[o]]))
names(x) <- names(o.list[[o]])
if (is.null(obj$altnames)) {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[o]])
}
else {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[o]],
unname(altnames[names(x)]))
}
}
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Biotype detection for sample ",n,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
categories=biotypes,
title=list(
text="Biotype",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
type="logarithmic",
showFirstLabel=FALSE,
min=1e-4,
tickInterval=1,
title=list(
useHTML=TRUE,
#text="Read count (log<sub>2</sub>)",
text="Expression (read count)",
margin=25,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
),
formatter=y.label.formatter
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
)
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
}
return(unquote_js_fun(json))
}
else if (by=="biotype") {
cols <- getColorScheme(length(samples))
box.list <- json <- vector("list",length(biotypes))
names(box.list) <- names(json) <- biotypes
for (b in biotypes) {
box.list[[b]] <- vector("list",length(samplenames))
names(box.list[[b]]) <- samplenames
for (n in samplenames)
box.list[[b]][[n]] <- counts[covars$biotype==b,n]
B <- boxplot(box.list[[b]],plot=FALSE)$stats
colnames(B) <- samplenames
o.list <- lapply(names(box.list[[b]]),function(x,M,b) {
v <- b[,x]
o <- which(M[[x]]<v[1] | M[[x]]>v[5])
if (length(o)>0)
return(M[[x]][o])
else
return(NULL)
},box.list[[b]],B)
names(o.list) <- samplenames
# Data series
BB <- matrix(0,nrow(B),ncol(B)) # Workaround of strange problem...
colnames(BB) <- colnames(B)
for (jj in 1:ncol(B))
BB[,jj] <- round(B[,jj],3)
d <- as.data.frame(BB)
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(samples))
names(series) <- names(samples)
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$name=s
if (grouped)
series[[s]]$color=cols$fill[counter]
else
series[[s]]$color=cols$fill[1]
m <- match(samples[[s]],colnames(d))
series[[s]]$data <- unname(as.list(d[,m]))
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(samples))
names(outliers) <- names(samples)
for (o in names(outliers)) {
counter <- counter + 1
outliers[[o]] <- list()
outliers[[o]]$id <- o
outliers[[o]]$name <- o
outliers[[o]]$type <- "scatter"
outliers[[o]]$showInLegend <- FALSE
if (grouped) {
outliers[[o]]$color <- cols$fill[counter]
outliers[[o]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
}
else {
outliers[[o]]$color <- cols$fill[1]
outliers[[o]]$marker <- list(
fillColor=cols$fill[1],
symbol="circle",
lineWidth=1,
lineColor=cols$border[1]
)
}
outliers[[o]]$data <- list()
m <- match(samples[[o]],colnames(d))
if (length(m)>0) {
for (i in m) {
x <- rep(d[1,i],length(o.list[[i]]))
names(x) <- names(o.list[[i]])
if (is.null(obj$altnames)) {
outliers[[o]]$data <-
make.highcharts.points(x,o.list[[i]])
}
else {
outliers[[o]]$data <- c(outliers[[o]]$data,
make.highcharts.points(x,o.list[[i]],
unname(altnames)))
}
}
}
}
json[[b]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Detection for biotype ",b,
" (population: ",lengths(box.list[[b]])[1],
")",sep="")
),
legend=list(
enabled=TRUE
),
xAxis=list(
categories=samplenames,
title=list(
text="Sample name",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
type="logarithmic",
showFirstLabel=FALSE,
min=1e-4,
tickInterval=1,
title=list(
text="Expression (read count)",
margin=25,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
),
formatter=y.label.formatter
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Condition {series.name}</span>',
'<br/>',
'<span style="font-weight:bold;">',
'Sample {point.key}',
'</span><br/>',sep=""
),
pointFormatter=tooltip.point.formatter
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Condition {series.name}</span>',
'<br/>',sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
}
return(unquote_js_fun(json))
}
}
#' Biodetection barplot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
bioDetectionToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
samples <- obj$samples
status <- obj$status
plotdata <- obj$user$plotdata
covars <- obj$user$covars
if (!is.null(samples)&& is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
names(plotdata$biotables) <- samplenames
}
# Otherwise we are using the names present in the input object
abu <- which(plotdata$genome>7)
nabu <- which(plotdata$genome<=7)
cols <- getColorScheme()
json <- vector("list",length(samplenames))
names(json) <- samplenames
for (n in samplenames) {
# Data series
series.abu <- vector("list",3)
names(series.abu) <- c("genome","detectionVSgenome","detectionVSsample")
series.abu$genome <- list()
series.abu$genome$id <- "abu_genome"
series.abu$genome$name <- "% in genome"
series.abu$genome$color <- cols$trans[1]
series.abu$genome$pointPlacement <- -0.2
series.abu$genome$data <- round(as.numeric(plotdata$genome[abu]),3)
series.abu$detectionVSgenome <- list()
series.abu$detectionVSgenome$id <- "abu_detected"
series.abu$detectionVSgenome$name <- "% detected"
series.abu$detectionVSgenome$color <- cols$trans[2]
series.abu$detectionVSgenome$pointPlacement <- 0
series.abu$detectionVSgenome$data <- round(as.numeric(
plotdata$biotables[[n]][1,abu]),3)
series.abu$detectionVSsample <- list()
series.abu$detectionVSsample$id <- "abu_sample"
series.abu$detectionVSsample$name <- "% in sample"
series.abu$detectionVSsample$color <- cols$trans[3]
series.abu$detectionVSsample$pointPlacement <- 0.2
series.abu$detectionVSsample$data <- round(as.numeric(
plotdata$biotables[[n]][2,abu]),3)
series.nabu <- vector("list",3)
names(series.nabu) <- c("genome","detectionVSgenome",
"detectionVSsample")
series.nabu$genome <- list()
series.nabu$genome$name <- "% in genome"
series.nabu$genome$yAxis <- 1
series.nabu$genome$pointStart <- length(abu)
series.nabu$genome$linkedTo <- "abu_genome"
series.nabu$genome$color <- cols$trans[1]
series.nabu$genome$pointPlacement <- -0.2
series.nabu$genome$data <- round(as.numeric(plotdata$genome[nabu]),3)
series.nabu$detectionVSgenome <- list()
series.nabu$detectionVSgenome$name <- "% detected"
series.nabu$detectionVSgenome$yAxis <- 1
series.nabu$detectionVSgenome$pointStart <- length(abu)
series.nabu$detectionVSgenome$linkedTo <- "abu_detected"
series.nabu$detectionVSgenome$color <- cols$trans[2]
series.nabu$detectionVSgenome$pointPlacement <- 0
series.nabu$detectionVSgenome$data <- round(as.numeric(
plotdata$biotables[[n]][1,nabu]),3)
series.nabu$detectionVSsample <- list()
series.nabu$detectionVSsample$name <- "% in sample"
series.nabu$detectionVSsample$yAxis <- 1
series.nabu$detectionVSsample$pointStart <- length(abu)
series.nabu$detectionVSsample$linkedTo <- "abu_sample"
series.nabu$detectionVSsample$color <- cols$trans[3]
series.nabu$detectionVSsample$pointPlacement <- 0.2
series.nabu$detectionVSsample$data <- round(as.numeric(
plotdata$biotables[[n]][2,nabu]),3)
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="column",
alignTicks=FALSE
),
title=list(
text=paste("Comparative biotype detection for ",
"sample ",n,sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
tooltip=list(
shared=TRUE
),
xAxis=list(
categories=names(plotdata$genome)[c(abu,nabu)],
title=list(
text="Biotype",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
),
plotLines=list(
list(
color="#8A8A8A",
width=1.5,
dashStyle="Dash",
value=length(abu)-0.5
)
),
plotBands=list(
list(
color="#FFFFE0",
from=-0.5,
to=length(abu)-0.5
),
list(
color="#FFECEB",
from=length(abu)-0.5,
to=length(plotdata$genome)
)
)
),
yAxis=list(
list(
min=0,
max=70,
title=list(
text="% of abundant features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
list(
min=0,
max=7,
title=list(
text="% of non-abundant features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
opposite=TRUE
)
),
plotOptions=list(
column=list(
grouping=FALSE,
shadow=FALSE,
groupPadding=0.3,
pointPadding=0.25,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'font-weight:bold">',
'{point.key}</span><br/>',sep=""
)
)
)
),
series=c(unname(series.abu),unname(series.nabu))
)
)
}
)
}
return(json)
}
#' Biotype saturation plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.noiseq}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.noiseq}}.
#' @param by Can be \code{"sample"} to create biotypes boxplots per sample or
#' \code{"biotype"} to create samples boxplots per biotype.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
bioSaturationToJSON <- function(obj,by=c("sample","biotype"),
jl=c("highcharts")) {
by <- tolower(by[1])
jl <- tolower(jl[1])
samples <- obj$samples
plotdata <- obj$user$plotdata
if (!is.null(samples)&& is.list(samples)) {
samplenames <- unlist(samples,use.names=FALSE)
names(plotdata) <- samplenames
}
if (by=="sample") {
json <- vector("list",length(samplenames))
names(json) <- samplenames
for (n in samplenames) {
depth <- round(plotdata[[n]][,1]/1e+6)
global <- round(plotdata[[n]][,2])
M <- plotdata[[n]][,3:ncol(plotdata[[n]])]
# To determine the separation
ord <- sort(M[nrow(M),],decreasing=TRUE,index.return=TRUE)
abu <- ord$ix[1:2]
names(abu) <- names(ord$x[1:2])
nabu <- ord$ix[3:length(ord$ix)]
names(nabu) <- names(ord$x[3:length(ord$x)])
cols <- getColorScheme(ncol(plotdata[[n]])-1)
counter <- 1
global.series <- list(
global=list(
id="global",
name="global",
color=cols$fill[counter],
data=make.highcharts.points(depth,global)
)
)
abu.series <- vector("list",2)
names(abu.series) <- names(abu)
for (s in names(abu.series)) {
counter <- counter + 1
abu.series[[s]] <- list()
abu.series[[s]]$id <- s
abu.series[[s]]$name <- s
abu.series[[s]]$color <- cols$fill[counter]
abu.series[[s]]$data <- make.highcharts.points(depth,
round(M[,s]))
}
nabu.series <- vector("list",length(3:ncol(M)))
names(nabu.series) <- names(nabu)
for (s in names(nabu.series)) {
counter <- counter + 1
nabu.series[[s]] <- list()
nabu.series[[s]]$id <- s
nabu.series[[s]]$name <- s
nabu.series[[s]]$color <- cols$fill[counter]
nabu.series[[s]]$data <- make.highcharts.points(depth,
round(M[,s]))
}
json[[n]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Biotype saturations for sample ",n,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
title=list(
text="Read depth (millions of reads)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
min=0,
max=max(global),
title=list(
text="Detected features",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
series=list(
lineWidth=2
),
scatter=list(
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Biotype {series.name}</span><br/>',
sep=""
),
pointFormat=paste(
"Depth: {point.x}M<br/>",
"Detected features: {point.y}",
sep="")
)
)
),
series=c(unname(global.series),
unname(abu.series),unname(nabu.series))
)
)
}
)
}
return(json)
}
else if (by=="biotype") {
biotypes <- colnames(plotdata[[1]])[2:ncol(plotdata[[1]])]
depths <- vector("list",length(plotdata))
names(depths) <- samplenames
for (n in samplenames)
depths[[n]] <- round(plotdata[[n]][,1]/1e+6)
json <- vector("list",length(biotypes))
names(json) <- biotypes
for (b in biotypes) {
series <- vector("list",length(plotdata))
names(series) <- samplenames
cols <- getColorScheme(length(samplenames))
counter <- 0
for (s in names(series)) {
counter <- counter + 1
series[[s]] <- list()
series[[s]]$id <- s
series[[s]]$name <- s
series[[s]]$color <- cols$fill[counter]
series[[s]]$data <- make.highcharts.points(depths[[s]],
round(plotdata[[s]][,b]))
}
json[[b]] <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Sample saturations for biotype ",b,
sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
xAxis=list(
title=list(
text="Read depth (millions of reads)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
title=list(
text="Detected features",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
series=list(
lineWidth=2
),
scatter=list(
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Sample {series.name}</span><br/>',
sep=""
),
pointFormat=paste(
"Depth: {point.x}M<br/>",
"Detected features: {point.y}",
sep="")
)
)
),
series=c(unname(series))
)
)
}
)
}
return(json)
}
}
#' Read noise plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.noiseq}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
readNoiseToJSON <- function(obj,jl=c("highcharts"),seed=42) {
jl <- tolower(jl[1])
d <- obj$user
samples <- obj$samples
# Too many points for a lot of curves of interactive data
if (nrow(d)>1000) {
set.seed(seed)
ii <- sort(sample(1:nrow(d),998))
ii <- c(1,ii,nrow(d))
d <- cbind(d[ii,1],d[ii,2:ncol(d)])
}
if (is.null(samples))
samples <- 1:(ncol(d)-1)
if (is.numeric(samples))
samplenames = colnames(dat)[samples+1]
if (is.list(samples))
samplenames <- unlist(samples)
cols <- getColorScheme(length(samplenames))
# Construct series
counter <- 0
series <- vector("list",length(samplenames))
names(series) <- samplenames
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
series[[n]]$color=cols$fill[counter]
series[[n]]$data <- make.highcharts.points(d[,1],d[,n])
series[[n]]$tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
)
}
switch(jl,
highcharts = {
json <- toJSON(list(
chart=list(
type="line",
zoomType="xy"
),
title=list(
text=paste("RNA-Seq mapped reads noise")
),
xAxis=list(
title=list(
text="% detected features",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=0,
max=100
),
yAxis=list(
title=list(
text="% of total reads",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
tickPositions=seq(0,110,10)
),
plotOptions=list(
line=list(
allowPointSelect=TRUE,
lineWidth=1,
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
),
turboThreshold=50000
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Boxplots JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.boxplot}}.
#'
#' @param obj A list holding boxplot data. See \code{\link{diaplot.boxplot}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
boxplotToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
b <- obj$plot
name <- obj$samples
status <- obj$status
altnames <- obj$altnames
o.list <- obj$user
grouped <- FALSE
if (is.null(name)) {
if (is.null(colnames(b$stats)))
nams <- paste("Sample",1:ncol(b$stats),sep=" ")
else
nams <- colnames(b$stats)
name <- list(Samples=nams)
}
else if (length(name)==1 && name=="none") {
nams <- rep("",ncol(b$stats))
name <- list(Samples=nams)
}
else if (is.list(name)) { # Is sample.list
nams <- unlist(name,use.names=FALSE)
grouped <- TRUE
}
cols <- getColorScheme()
# Data series
d <- as.data.frame(round(b$stat,3))
ids <- 0:(ncol(d)-1)
d <- rbind(ids,d)
colnames(d) <- nams
names(ids) <- colnames(d)
counter <- 0
series <- vector("list",length(name))
names(series) <- names(name)
for (n in names(series)) {
counter <- counter + 1
series[[n]] <- list()
series[[n]]$name=n
if (grouped)
series[[n]]$color=cols$fill[counter]
else
series[[n]]$color=cols$fill[1]
m <- match(name[[n]],colnames(d))
series[[n]]$data <- unname(as.list(d[,m]))
}
# Outlier series (if any)
counter <- 0
outliers <- vector("list",length(name))
names(outliers) <- names(name)
for (n in names(outliers)) {
counter <- counter + 1
outliers[[n]] <- list()
outliers[[n]]$id <- n
outliers[[n]]$name <- n
outliers[[n]]$type <- "scatter"
outliers[[n]]$showInLegend <- FALSE
if (grouped) {
outliers[[n]]$color <- cols$fill[counter]
outliers[[n]]$marker <- list(
fillColor=cols$fill[counter],
symbol="circle",
lineWidth=1,
lineColor=cols$border[counter]
)
}
else {
outliers[[n]]$color <- cols$fill[1]
outliers[[n]]$marker <- list(
fillColor=cols$fill[1],
symbol="circle",
lineWidth=1,
lineColor=cols$border[1]
)
}
outliers[[n]]$data <- list()
m <- match(name[[n]],colnames(d))
if (length(m)>0) {
for (i in m)
x <- rep(d[1,i],length(o.list[[i]]))
names(x) <- names(o.list[[i]])
outliers[[n]]$data <- c(outliers[[n]]$data,
make.highcharts.points(x,o.list[[i]],unname(altnames)))
}
}
# Boxplot tooltip point formatter for the case of zeros
tooltip.point.formatter <- paste("function() {",
" var min = this.low === 0.001 ? 0 : this.low;" ,
" var q1 = this.q1 === 0.001 ? 0 : this.q1;" ,
" var med = this.median === 0.001 ? 0 : this.median;",
" var q3 = this.q3 === 0.001 ? 0 : this.q3;",
" var max = this.high === 0.001 ? 0 : this.high;",
" var str = 'Maximum: ' + max + '<br/>' +",
" 'Upper quartile: ' + q3 + '<br/>' +",
" 'Median: ' + med + '<br/>' +",
" 'Lower quartile: ' + q1 + '<br/>' +",
" 'Minimum: ' + min + '<br/>';",
" return str;",
"}",sep="")
# Legend clicker
boxplot.onclick <- paste("function() {",
" var chart = this.chart;",
" var outlier_id = chart.get(this.name);",
" if (!outlier_id.visible) {",
" outlier_id.show();",
" } else {",
" outlier_id.hide();",
" }",
"}",sep="")
if (is.null(obj$altnames)) {
outlier.pointformat=paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
else {
outlier.pointformat=paste(
'<strong>Sample {point.category}</strong><br/>',
'Gene ID: {point.name}<br/>',
'Gene name: {point.alt_name}<br/>',
'Value: {point.y}<br/>',sep=""
)
}
json <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="boxplot"
),
title=list(
text=paste("Boxplot ",status,sep="")
),
legend=list(
enabled=TRUE
),
xAxis=list(
categories=nams,
title=list(
text="Sample name",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Read count (log<sub>2</sub>)",
margin=30,
style=list(
color="#000000",
fontSize="1.1em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
)
),
plotOptions=list(
boxplot=list(
fillColor="#F0F0E0",
lineWidth=2,
medianColor="#000000",
medianWidth=3,
stemColor="#000000",
stemDashStyle="dash",
stemWidth=1,
whiskerColor="#000000",
whiskerLength="75%",
whiskerWidth=1,
grouping=FALSE,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'color:{series.color};',
'font-weight:bold">',
'\u25CF </span>',
'<span style="font-size:1.1em;',
'font-weight:bold">',
'Condition {series.name}</span><br/>',
'<span style="font-weight:bold">',
'Sample {point.key}</span><br/>',sep=""
),
pointFormatter=tooltip.point.formatter
),
events=list(
legendItemClick=boxplot.onclick
)
),
scatter=list(
allowPointSelect=TRUE,
tooltip=list(
headerFormat=paste(
'<span style="font-weight:bold;',
'color:{series.color};">',
'\u25CF </span>',
'<span style="font-weight:bold">',
'Condition {series.name}</span><br/>',
sep=""
),
pointFormat=outlier.pointformat
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
)
)
),
series=c(unname(series),unname(outliers))
)
)
}
)
return(unquote_js_fun(json))
}
#' GC/length bias plot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.edaseq}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.noiseq}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
biasPlotToJSON <- function(obj,jl=c("highcharts"),seed=1) {
jl <- tolower(jl[1])
counts <- round(nat2log(obj$user$counts),3)
status <- obj$status
covar <- obj$user$covar
covarname <- obj$user$covarname
samples <- obj$samples
# Too many points for a lot of curves of interactive data
if (nrow(counts)>2000) {
set.seed(seed)
ii <- sample(1:nrow(counts),2000)
counts <- counts[ii,]
covar <- covar[ii]
}
# If length bias, not nice to have x-axis at -200k
min.x <- ifelse(max(covar>100),0,"undefined")
if (is.null(samples)) {
if (is.null(colnames(x)))
samplenames <- paste("Sample",1:ncol(counts),sep=" ")
else
samplenames <- colnames(counts)
samples <- list(Samples=nams)
cols <- getColorScheme(length(samples))
}
else if (is.list(samples)) { # Is sample.list
samplenames <- unlist(samples,use.names=FALSE)
grouped <- TRUE
}
colnames(counts) <- samplenames
# Construct series
counter <- 0
series <- vector("list",length(samplenames))
names(series) <- samplenames
for (n in names(series)) {
counter <- counter + 1
x <- counts[,n]
fit <- lowess(covar,x)
series[[n]] <- list()
series[[n]]$name <- n
series[[n]]$color <- cols$fill[counter]
series[[n]]$data <- lapply(1:length(x),function(i,x,y) {
return(c(x[i],y[i])) },round(fit$x,3),round(fit$y,3))
}
switch(jl,
highcharts = {
json <- toJSON(list(
chart=list(
type="line",
zoomType="xy"
),
title=list(
text=paste(covarname," bias detection - ",status)
),
xAxis=list(
min=min.x,
title=list(
text=covarname,
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Read count (log<sub>2</sub>)",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE
),
plotOptions=list(
line=list(
marker=list(
enabled=FALSE,
states=list(
hover=list(
enabled=FALSE
)
)
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=NULL
),
turboThreshold=50000
)
),
series=unname(series)
)
)
}
)
return(json)
}
#' Filtered genes barplot JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diaplot.filtered}}.
#'
#' @param obj A list holding plot data. See \code{\link{diaplot.filered}}.
#' @param by Either \code{"chromosome"} or \code{"biotype"}
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
filteredToJSON <- function(obj,by=c("chromosome","biotype"),
jl=c("highcharts")) {
jl <- tolower(jl[1])
by <- tolower(by[1])
filtered <- obj$user$filtered
total <- obj$user$total
cols <- getColorScheme(2)
if (by=="chromosome") {
chr <- table(as.character(filtered$chromosome))
chr.all <- table(as.character(total$chromosome))
barlab.chr <- as.character(chr)
per.chr <- round(chr/chr.all[names(chr)],3)
per.chr[per.chr>1] <- 1
series <- vector("list",2)
names(series) <- c("number","fraction")
series$number <- list()
series$number$id <- "chr_number"
series$number$name <- "Number of genes"
series$number$color <- cols$fill[1]
series$number$pointPlacement <- -0.2
series$number$data <- unname(chr)
series$fraction <- list()
series$fraction$id <- "chr_fraction"
series$fraction$name <- "Fraction of total genes"
series$fraction$color <- cols$fill[2]
series$fraction$pointPlacement <- 0.2
series$fraction$yAxis <- 1
series$fraction$data <- unname(per.chr)
what <- chr
}
else if (by=="biotype") {
bt <- table(as.character(filtered$biotype))
bt.all <- table(as.character(total$biotype))
barlab.bt <- as.character(bt)
per.bt <- round(bt/bt.all[names(bt)],3)
per.bt[per.bt>1] <- 1
series <- vector("list",2)
names(series) <- c("number","fraction")
series$number <- list()
series$number$id <- "bt_number"
series$number$name <- "Number of genes"
series$number$color <- cols$fill[1]
series$number$pointPlacement <- -0.2
series$number$data <- unname(bt)
series$fraction <- list()
series$fraction$id <- "bt_fraction"
series$fraction$name <- "Fraction of total genes"
series$fraction$color <- cols$fill[2]
series$fraction$pointPlacement <- 0.2
series$fraction$yAxis <- 1
series$fraction$data <- unname(per.bt)
what <- bt
}
json <- switch(jl,
highcharts = {
toJSON(
list(
chart=list(
type="column",
alignTicks=FALSE
),
title=list(
text=paste("Filtered genes per ",by,sep="")
),
legend=list(
enabled=TRUE,
itemHoverStyle=list(
color="#B40000"
)
),
tooltip=list(
shared=TRUE
),
xAxis=list(
categories=names(what),
title=list(
text=paste(toupper(substr(by,1,1)),substr(by,2,
nchar(by)),sep=""),
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontWeight="bold"
)
)
),
yAxis=list(
list(
lineColor=cols$fill[1],
lineWidth=2,
min=0,
tickAmount=11,
title=list(
text="Number of genes",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
offset=10
),
list(
lineColor=cols$fill[2],
lineWidth=2,
min=0,
max=1,
tickAmount=11,
#tickInterval=0.1,
title=list(
text="Fraction of total genes",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
opposite=TRUE,
offset=10
)
),
plotOptions=list(
column=list(
grouping=FALSE,
shadow=FALSE,
groupPadding=0.2,
pointPadding=0.2,
tooltip=list(
headerFormat=paste(
'<span style="font-size:1.1em;',
'font-weight:bold">',
'{point.key}</span><br/>',sep=""
)
)
)
),
series=c(unname(series))
)
)
}
)
return(json)
}
#' Volcano JSON exporter for the metaseqR package
#'
#' Non-exportable JSON exporter for \code{\link{diagplot.volcano}}.
#'
#' @param obj A list holding volcano plot data. See \code{\link{diaplot.volcano}}.
#' @param jl JavaScript charting library to export. Currently only \code{"highcharts"}
#' supported.
#' @return A JSON string.
#' @author Panagiotis Moulos
volcanoToJSON <- function(obj,jl=c("highcharts")) {
jl <- tolower(jl[1])
f <- obj$x
p <- obj$y
xlim <- obj$xlim
ylim <- obj$ylim
alt.names <- obj$altnames
pcut <- obj$pcut
fcut <- obj$fcut
up <- obj$user$up
down <- obj$user$down
ff <- obj$user$unf
pp <- obj$user$unp
alt.names.neutral <- obj$user$ualt
con <- obj$user$con
switch(jl,
highcharts = {
if (is.null(alt.names))
point.format=paste("<strong>Gene ID: </strong>{point.name}<br>",
"<strong>Fold change: </strong>{point.x}<br>",
"<strong>Significance: </strong>{point.y}",sep="")
else
point.format=paste("<strong>Gene name: </strong>",
"{point.alt_name}<br>",
"<strong>Gene ID: </strong>{point.name}<br>",
"<strong>Fold change: </strong>{point.x}<br>",
"<strong>Significance: </strong>{point.y}",sep="")
json <- toJSON(
list(
chart=list(
type="scatter",
zoomType="xy"
),
title=list(
text=paste("Volcano plot for",con)
),
xAxis=list(
title=list(
text="Fold change",
margin=20,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(xlim[1],3),
max=round(xlim[2],3)
),
yAxis=list(
title=list(
useHTML=TRUE,
text="Significance (-log<sub>10</sub>(p-value))",
margin=25,
style=list(
color="#000000",
fontSize="1.2em"
)
),
labels=list(
style=list(
color="#000000",
fontSize="1.1em",
fontWeight="bold"
)
),
startOnTick=TRUE,
endOnTick=TRUE,
showLastLabel=TRUE,
gridLineWidth=1,
min=round(ylim[1]-2,3),
max=round(ylim[2],3)
),
#legend=list(
# layout="vertical",
# align="left",
# verticalAlign="top",
# floating=TRUE,
# backgroundColor="#FFFFFF",
# borderWidth=1
#),
plotOptions=list(
scatter=list(
allowPointSelect=TRUE,
marker=list(
radius=2,
states=list(
hover=list(
enabled=TRUE,
lineColor="#333333"
)
)
),
states=list(
hover=list(
marker=list(
enabled=FALSE
)
)
),
tooltip=list(
headerFormat=paste("<span style=",
"\"font-size:1.1em;color:{series.color};",
"font-weight:bold\">{series.name}<br>",
sep=""),
pointFormat=point.format
),
turboThreshold=50000
)
),
series=list(
list(
name="Up-regulated",
color="#EE0000",
marker=list(
symbol="circle"
),
data=make.highcharts.points(f[up],-log10(p[up]),
unname(alt.names[up]))
),
list(
name="Down-regulated",
marker=list(
symbol="circle"
),
color="#00CD00",
data=make.highcharts.points(f[down],-log10(p[down]),
unname(alt.names[down]))
),
list(
name="Unregulated",
marker=list(
symbol="circle"
),
color="#0000EE",
data=make.highcharts.points(ff,-log10(pp),
unname(alt.names.neutral))
),
list(
name="Downfold threshold",
color="#000000",
type="line",
dashStyle="dash",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.x}<br/>',sep="")
),
data=list(round(c(-fcut,ylim[1]-5),3),
round(c(-fcut,ylim[2]),3))
),
list(
name="Upfold threshold",
color="#000000",
type="line",
dashStyle="Dash",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.x}<br/>',sep="")
),
data=list(round(c(fcut,ylim[1]-5),3),
round(c(fcut,ylim[2]),3))
),
list(
name="Significance threshold",
color="#000000",
type="line",
dashStyle="DashDot",
marker=list(
enabled=FALSE
),
tooltip=list(
headerFormat=paste('<strong>{series.name}',
'</strong><br/>',sep=""),
pointFormat=paste('<strong>Threshold: ',
'</strong>{point.y}<br/>',sep="")
),
data=list(round(c(xlim[1],-log10(pcut)),3),
round(c(xlim[2],-log10(pcut)),3))
)
)
)
)
}
)
return(json)
}
unquote_js_fun <- function(js) {
if (is.list(js))
js <- lapply(js,unquote_js_fun)
else {
op <- gregexpr(pattern="function",js)
cl <- gregexpr(pattern="}\\\"",js)
if (length(op)>0) {
starts <- as.numeric(op[[1]])
for (i in 1:length(starts))
substr(js,starts[i]-1,starts[i]-1) <- " "
ends <- as.numeric(cl[[1]])
for (i in 1:length(starts))
substr(js,ends[i]+1,ends[i]+1) <- " "
}
}
return(js)
}
getGroupColorScheme <- function(group) {
cols <- getColorScheme(length(group))
classes <- as.factor(as.class.vector(group))
design <- as.numeric(classes)
return(lapply(cols,function(x,classes,design) {
return(x[1:length(levels(classes))][design])
},classes,design))
}
getColorScheme <- function(n=NULL) {
if (missing(n) || is.null(n))
return(getColors())
else {
cols <- getColors()
if (n > length(cols$fill)) {
cols$fill <- rep(cols$fill,length.out=n)
cols$border <- rep(cols$border,length.out=n)
cols$select <- rep(cols$select,length.out=n)
cols$trans <- rep(cols$trans,length.out=n)
}
return(cols)
}
}
getColors <- function() {
return(list(
fill=c("#CD0000","#00CD00","#0000EE","#FFD700","#87CEEB","#CD8500",
"#DEB887","#FF0000","#0000FF","#00FF00","#FFA500","#A9A9A9",
"#008B00","#313131","#FFC0CB","#A52A2A","#FF00FF","#9ACD32",
"#8B636C","#2E8B57","#008B8B"),
border=c("#850000","#006B00","#000085","#927C00","#156280","#5A3A00",
"#8B7457","#935E18","#000080","#008500","#603E00","#454545",
"#073E07","#000000","#896067","#691111","#7C007C","#3A4D14",
"#5B1726","#0C2517","#062A2A"),
selected=c("#FF0000","#00FF00","#0066FF","#FFD700","#FFEB77","#FFB428",
"#FFD9A5","#FF326D","#0089FF","#B3FF00","#FFC352","#D9D9D9",
"#00EC00","#8E8E8E","#FFDAE0","#F94444","#FF87FF","#C2FF45",
"#EA889D","#4EE590","#00DADA"),
trans=c("rgba(205,0,0,0.6)","rgba(0,205,0,0.6)","rgba(0,0,238,0.6)",
"rgba(255,215,0,0.6)","rgba(135,206,235,0.6)","rgba(205,133,0,0.6)",
"rgba(222,184,135,0.6)","rbga(255,0,0,0.5)","rgba(0,0,255,0.5)",
"rgba(0,255,0,0.5)","rgba(255,165,0,0.6)","rgba(169,169,169,0.5)",
"rgba(0,139,0,0.6)","rgba(49,49,49,0.6)","rgba(255,192,203,0.5)",
"rgba(165,42,42,0.6)","rgba(255,0,255,0.6)","rgba(154,205,50,0.6)",
"rgba(139,99,108,0.6)","rgba(46,139,87,0.6)","rgba(0,139,139,0.6)")
))
}
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.