R/noiseq.R
In pmoulos/metaseqR2-local: An R package for the analysis and result reporting of RNA-Seq data by combining multiple statistical algorithms
Defines functions
addData
readData
PCA.GENES
.calcFactorWeighted
.calcFactorQuantile
.calcNormFactors
make.ASCA.design
filtered.data
CV
mypretty
DE.plot
dat
countsbio.plot
biodetection.plot
biodetection.dat
logscaling
plot.y2
ranking
sim.samples
int.mult
ASCAfun.triple
ASCAfun2
ASCAfun1
ASCA.3f
ASCA.2f
ASCA.1f
ARSyNseq
ARSyNmodel
ARSyNcomponents
allMDbio
################################################################################
# classes.R
setClass("Biodetection",representation(dat="list"))
setClass("CD",representation(dat="list"))
setClass("CountsBio",representation(dat="list"))
setClass("GCbias",representation(dat="list"))
setClass("lengthbias",representation(dat="list"))
setClass("Saturation",representation(dat="list"))
setClass("PCA",representation(dat="list"))
setGeneric("explo.plot",function(object,...) standardGeneric("explo.plot"))
setMethod("explo.plot","Biodetection",
function(object,samples=c(1,2),plottype=c("persample","comparison"),
toplot="protein_coding",...)
biodetection.plot(object@dat,samples=samples,plottype=plottype,
toplot=toplot,...))
setMethod("explo.plot","CD",function(object,samples=NULL,...)
cd.plot(object@dat,samples=samples,...))
setMethod("explo.plot","CountsBio",
function(object,samples=c(1,2),toplot="global",
plottype=c("barplot","boxplot"),...)
countsbio.plot(object@dat,samples,toplot,plottype,...))
#setMethod("explo.plot","GCbias",
# function(object,samples=NULL,toplot="global",...)
# GC.plot(object@dat,samples=samples,toplot=toplot,...))
#setMethod("explo.plot","lengthbias",
# function(object,samples=NULL,toplot="global",...)
# length.plot(object@dat,samples=samples,toplot=toplot,...))
#setMethod("explo.plot","Saturation",
# function(object,samples=NULL,toplot=1,yleftlim=NULL,yrightlim=NULL,...)
# saturation.plot(object@dat,samples=samples,toplot=toplot,yleftlim=yleftlim,
# yrightlim=yrightlim,...))
#setMethod("explo.plot","PCA",
# function(object,samples=c(1,2),plottype="scores",factor=NULL)
# PCA.plot(object@dat,samples=samples,plottype=plottype,factor=factor))
# Show methods for exploration objects
setMethod("show","Biodetection",function(object) {
cat("\n Reference genome: \n==========\n")
names(dimnames(object@dat$genome))=NULL
print(object@dat$genome)
for (i in seq_len(length(object@dat$biotables))) {
cat("\n",names(object@dat$biotables)[i],"\n==========\n")
print(object@dat$biotables[[i]])
}
})
setMethod("show","CD",function(object) {
cat("\n Confidence intervals for median of M to compare each sample ",
"to reference:\n=======\n")
print(object@dat$DiagnosticTest)
cat("\n Reference sample is:\n=======\n")
print(object@dat$refColumn)
})
setMethod("show","CountsBio",function(object) {
cat("\n Summary: \n============\n")
print(object@dat$summary[[1]])
})
setMethod("show","GCbias",function(object) {
x <- object@dat$RegressionModels
for (i in seq_len(length(x))) {
print(names(x)[i])
print(summary(x[[i]]))
}
})
setMethod("show","lengthbias",function(object) {
x <- object@dat$RegressionModels
for (i in seq_len(length(x))) {
print(names(x)[i])
print(summary(x[[i]]))
}
})
setMethod("show","Saturation",function(object) {
x <- dat2save(object)
cat("\n Number of detected features at each sequencing ",
"depth: \n============\n")
for (i in seq_len(length(x))) {
print(names(x)[i])
print(x[[i]])
}
})
setMethod("show","PCA",function(object) {
x <- object$result$var.exp
x <- round(x*100,4)
colnames(x) <- c("%Var","Cum %Var")
rownames(x) <- paste("PC",seq_len(nrow(x)))
cat("\n Percentage of total variance explained by each component: ",
"\n============\n")
print(x)
})
# Coercion methods for exploration objects
setGeneric("dat2save",function(object) standardGeneric("dat2save"))
setMethod("dat2save","Biodetection",function(object) object@dat)
setMethod("dat2save","CD",function(object) object@dat)
setMethod("dat2save","CountsBio",function(object) object@dat$summary)
setMethod("dat2save","GCbias",function(object) object@dat$data2plot)
setMethod("dat2save","lengthbias",function(object) object@dat$data2plot)
setMethod("dat2save","Saturation",function(object) {
muestras=vector("list",length=length(object@dat$depth))
names(muestras)=names(object@dat$depth)
for (i in seq_len(length(muestras))) {
muestras[[i]] <- object@dat$depth[[i]]
for (j in seq_len(length(object@dat$saturation))) {
muestras[[i]] <- cbind(muestras[[i]],
object@dat$saturation[[j]][[i]])
}
colnames(muestras[[i]]) <- c("depth",names(object@dat$saturation))
}
muestras
})
setMethod("dat2save","PCA",function(object) object@dat$result)
setClass("myInfo",representation(method="character",k="numeric",
lc="numeric",factor="vector",v="numeric",nss="numeric",pnr="numeric",
comparison="vector",replicates="character"))
setClass("Output",representation(results="list"),contains="myInfo")
setValidity("Output",function(object) {
if (!(is.character(object@method))) {
return(paste("Method must be a string"))
} else if (!(is.numeric(object@k))) {
return(paste("k must be numeric"))
} else if (!(is.numeric(object@lc))) {
return(paste("lc must be numeric"))
} else if (!(is.vector(object@factor))) {
return(paste("Factor must be a vector of strings"))
} else if (!(is.numeric(object@v))) {
return(paste("v must be numeric"))
} else if (!(is.numeric(object@nss))) {
return(paste("nss must be numeric"))
} else if (!(is.numeric(object@pnr))) {
return(paste("pnr must be numeric"))
} else if (!(is.vector(object@comparison))) {
return(paste("Comparison must be a vector of strings"))
} else if (!(is.list(object@results))) {
return(paste("Results must be a list of data.frames"))
} else {
return(TRUE)
}
})
Output <- function (data,method,k,lc,factor,v,nss,pnr,comparison,replicates) {
new("Output",results=data,method=method,k=k,lc=lc,factor=factor,v=v,nss=nss,
pnr=pnr,comparison=comparison,replicates=replicates)
}
setMethod("show","Output",function(object) {
if (object@method == "n")
object@method="none"
for (i in seq_len(length(object@results))) {
cat("\nSummary",i,"\n=========\n")
cat("\nYou are comparing",object@comparison[i],"from",object@factor[i],
"\n\n")
print(head(object@results[[i]][order(object@results[[i]][,5],
decreasing=TRUE),]))
}
cat("\nNormalization\n")
cat("\tmethod:",object@method,"\n")
cat("\tk:",object@k,"\n")
cat("\tlc:",object@lc,"\n")
# Simulated samples
if (object@replicates == "no") {
cat("\nYou are working with simulated replicates:\n")
cat("\tpnr:",object@pnr,"\n")
cat("\tnss:",object@nss,"\n")
cat("\tv:",object@v,"\n")
}
# With biological or technical replicates
else {
cat("\nYou are working with",object@replicates,"replicates\n")
}
})
################################################################################
################################################################################
# allMD.R
allMD <- function (input,factor,conditions,k=0.5,replicates,norm="rpkm",pnr=0.2,
nss=5,v=0.02,lc=0) {
# Check if the factor introduced is already defined
# If the factor introduced is defined and has more than 2 conditions,it will
# check if the conditions specified are defined too
condition_fac <- FALSE
condition_lev <- FALSE
datos1 <- datos2 <- matrix()
for (i in colnames(pData(input))) {
if (factor == i) {
condition_fac <- TRUE
if (!is.factor(pData(input)[,i]))
pData(input)[,i]=as.factor(pData(input)[,i])
if (length(levels(pData(input)[,i])) == 2) {
if (!is.null(assayData(input)$exprs)) {
datos1 <-
assayData(input)$exprs[,which(
pData(input)[,i]==levels(pData(input)[,i])[1]),
drop=FALSE]
datos2 <-
assayData(input)$exprs[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2]),
drop=FALSE]
}
else {
datos1 <-
assayData(input)$counts[,which(
pData(input)[,i]==levels(pData(input)[,i])[1]),
drop=FALSE]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i]==levels(pData(input)[,i])[2]),
drop=FALSE]
}
# Define the comparison string
comparison <- paste(levels(pData(input)[,i])[1],
levels(pData(input)[,i])[2],sep=" - ")
condition_lev <- TRUE
}
else {
if (is.null(conditions))
stop("Error. You must specify which conditions you wish ",
"to compare when the factor has two or more ",
"conditions.\n")
if (length(conditions) != 2)
stop("Error. The argument conditions must contain the 2 ",
"conditions you wish to compare.")
l <- conditions %in% pData(input)[,i]
# If they are defined,they will be TRUE
if (l[1] == TRUE && l[2] == TRUE) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[1]),drop=FALSE]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[2]),drop=FALSE]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[1]),drop=FALSE]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[2]),drop=FALSE]
}
# Define the comparison string
comparison <- paste(conditions[1],conditions[2],sep=" - ")
condition_lev <- TRUE
}
}
}
}
if (!condition_fac)
stop("The factor you have written does not correspond with any of the ",
"ones you have defined.")
if (!condition_lev)
stop("The conditions you have written don't exist in the factor ",
"specified.\n")
# Correction to make it work when there are simulated samples
if (replicates == "no")
replicates <- "technical"
n1 <- ncol(as.matrix(datos1))
n2 <- ncol(as.matrix(datos2))
if (norm == "n") { # no normalization
datos1 <- round(datos1,100)
datos2 <- round(datos2,100)
}
if (is.null(k)) {
m1 <- min(datos1[noceros(datos1,num=FALSE)],na.rm=TRUE)
m2 <- min(datos2[noceros(datos2,num=FALSE)],na.rm=TRUE)
mm <- min(m1,m2)
k <- mm/2
}
# Total counts for each gene:
suma1 <- rowSums(as.matrix(datos1))
suma2 <- rowSums(as.matrix(datos2))
# All genes
todos <- rownames(as.matrix(datos1))
# Genes with counts in any condition
concounts <- names(which(suma1+suma2 > 0))
long <- 1000
g.sinL <- NULL
if (!is.null(featureData(input)@data$Length)) {
g.sinL <- names(which(is.na(featureData(input)@data$Length)))
if (any(!is.na(featureData(input)@data$Length)) == TRUE)
long <- featureData(input)@data[concounts,"Length"]
}
if (replicates == "technical") { ### technical replicates
suma1 <- suma1[concounts]
suma2 <- suma2[concounts]
#----------------------------------------------------------------------#
# Normalization of counts for each condition (aggregating replicates)
if (norm == "rpkm") { # RPKM
suma1.norm <- noirpkm(suma1,long=long,k=k,lc=lc)
suma2.norm <- noirpkm(suma2,long=long,k=k,lc=lc)
}
if (norm == "uqua") {
suma.norm <- noiuqua(cbind(suma1,suma2),long=long,lc=lc,k=k)
suma1.norm <- as.matrix(suma.norm[ ,1])
suma2.norm <- as.matrix(suma.norm[ ,2])
}
if (norm == "tmm") {
suma.norm <- noitmm(as.matrix(cbind(suma1,suma2)),long=long,lc=lc,
k=k)
suma1.norm <- as.matrix(suma.norm[ ,1])
suma2.norm <- as.matrix(suma.norm[ ,2])
}
}
#-------------------------------------------------------------------------#
## Noise distribution
if ((n1+n2)>2) { # with real samples
datitos <- cbind(datos1,datos2)
datitos <- datitos[concounts,]
gens.sin0 <- setdiff(concounts,g.sinL)
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
datos1.norm <- datitos.norm[ ,seq_len(n1)]
datos2.norm <- datitos.norm[ ,(n1+1):(n1+n2)]
if (n1 > 1) {
MD1 <- MD(dat=datos1.norm)
}
else {
MD1 <- NULL
}
if (n2 > 1) {
MD2 <- MD(dat=datos2.norm)
}
else {
MD2 <- NULL
}
}
else { # with simulated samples
if (nss == 0) {
nss <- 5
}
datos.sim <- sim.samples(counts1=sinceros(suma1,k=k),
counts2=sinceros(suma2,k=k),pnr=pnr,nss=nss,v=v)
nn <- vapply(datos.sim,ncol,numeric(1))
dat.sim.norm <- vector("list",length=2)
datitos <- cbind(datos.sim[[1]],datos.sim[[2]])
rownames(datitos)=names(suma1)
sumita <- rowSums(datitos)
g.sin0 <- names(which(sumita > 0))
gens.sin0 <- setdiff(g.sin0,g.sinL)
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
dat.sim.norm[[1]] <- datitos.norm[,seq_len(nn[1])]
dat.sim.norm[[2]] <- datitos.norm[,(nn[1]+1):sum(nn)]
MD1 <- MD(dat=dat.sim.norm[[1]])
MD2 <- MD(dat=dat.sim.norm[[2]])
}
Mr <- c(as.numeric(MD1$M),as.numeric(MD2$M))
Dr <- c(as.numeric(MD1$D),as.numeric(MD2$D))
#-------------------------------------------------------------------------#
## M and D for different experimental conditions
if (replicates == "technical" & norm != "n") {
MDs <- MD(dat=cbind(suma1.norm,suma2.norm))
lev1 <- suma1.norm[,1]
lev1 <- lev1[todos]
lev2 <- suma2.norm[,1]
lev2 <- lev2[todos]
}
else {
if ((n1+n1) == 2) {
datos1.norm <- sinceros(as.matrix(datos1)[concounts,],k=k)
datos2.norm <- sinceros(as.matrix(datos2)[concounts,],k=k)
}
resum1.norm <- rowMeans(as.matrix(datos1.norm))
resum2.norm <- rowMeans(as.matrix(datos2.norm))
lev1 <- resum1.norm[todos]
lev2 <- resum2.norm[todos]
MDs <- MD(dat=cbind(resum1.norm,resum2.norm))
}
## Completing M and D
names(lev1) <- names(lev2) <- todos
Ms <- as.numeric(MDs$M)
names(Ms) <- rownames(MDs$M)
Ms <- Ms[todos]
names(Ms) <- todos
Ds <- as.numeric(MDs$D)
names(Ds) <- rownames(MDs$D)
Ds <- Ds[todos]
names(Ds) <- todos
## Results
list("k"=k,"comp"=comparison,"Level1"=lev1,"Level2"=lev2,"Ms"=Ms,"Ds"=Ds,
"Mn"=Mr,"Dn"=Dr)
}
allMDbio <- function(input,factor,conditions,k=0.5,norm="rpkm",lc=1,r=10,
a0per=0.9,nclust=15,filter=1,depth=NULL,cv.cutoff=0,cpm=1) {
# Check if the factor introduced is already defined
# If the factor introduced is defined and has more than 2 conditions,
# it will check if the conditions specified are defined too
condition_fac <- FALSE
condition_lev <- FALSE
datos1 <- datos2 <- matrix()
for (i in colnames(pData(input))) {
if (factor == i) {
condition_fac <- TRUE
if (!is.factor(pData(input)[,i]))
pData(input)[,i] <- as.factor(pData(input)[,i])
if (length(levels(pData(input)[,i])) == 2) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i]==levels(pData(input)[,i])[1])]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2])]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] ==levels(pData(input)[,i])[1])]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2])]
}
# Define the comparison string
comparison <- paste(levels(pData(input)[,i])[1],
levels(pData(input)[,i])[2],sep=" - ")
condition_lev <- TRUE
if (!((ncol(datos1) > 1) && (ncol(datos2) > 1)))
stop("Error. NOISeqBIO needs at least 2 biological replicates ",
"per condition.\n")
}
else {
if (is.null(conditions))
stop("Error. You must specify which conditions you wish ",
"to compare when the factor has two or more ",
"conditions.\n")
if (length(conditions) != 2)
stop("Error. The argument conditions must contain the 2 ",
"conditions you wish to compare.")
l <- conditions %in% pData(input)[,i]
# If they are defined,they will be TRUE
if (l[1] == TRUE && l[2] == TRUE) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[1])]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[2])]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[1])]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[2])]
}
# Define the comparison string
comparison <- paste(conditions[1],conditions[2],sep=" - ")
condition_lev <- TRUE
}
}
}
}
if (condition_fac == FALSE)
stop("The factor specified does not correspond with any of the ",
"ones you have defined.")
if (condition_lev == FALSE)
stop("The conditions specified don't exist for the factor specified.\n")
#--------------------------------------------------------------------------#
# Number of observations within each condition
n1 <- ncol(as.matrix(datos1))
n2 <- ncol(as.matrix(datos2))
if (max(n1,n2) == 1)
stop("There is only one replicate per condition. Please,use ",
"NOISeq instead of NOISeqBIO.\n")
# Rounding off data
if (norm == "n") { # no normalization
datos1 <- round(datos1,10)
datos2 <- round(datos2,10)
}
# Computing k
if (is.null(k)) {
m1 <- min(datos1[noceros(datos1,num=FALSE)],na.rm=TRUE)
m2 <- min(datos2[noceros(datos2,num=FALSE)],na.rm=TRUE)
k <- min(m1,m2)/2
}
# Total counts for each gene:
suma1 <- rowSums(as.matrix(datos1))
suma2 <- rowSums(as.matrix(datos2))
# Genes with counts in any condition
concounts <- names(which(suma1+suma2 > 0))
# All genes
todos <- rownames(as.matrix(datos1))
# Gene length
long <- 1000
g.sinL <- NULL # genes with no length defined
if (!is.null(featureData(input)@data$Length)) {
g.sinL <- names(which(is.na(featureData(input)@data$Length)))
if (any(!is.na(featureData(input)@data$Length)) == TRUE)
long <- featureData(input)@data[concounts,"Length"]
}
# Genes with counts and with length
gens.sin0 <- setdiff(concounts,g.sinL)
# cond1 and cond2 in the same matrix
datitos <- cbind(datos1,datos2)
datitos <- datitos[concounts,] # selecting only genes with counts
# Sequencing depth when filtering method=3
if (filter == 3 && is.null(depth))
depth <- colSums(datitos)
#-------------------------------------------------------------------------#
#-------------------------------------------------------------------------#
## Normalization
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
#-------------------------------------------------------------------------#
## Filtering out low count features
if (filter != 0) {
datos.filt <- filtered.data(dataset=datitos.norm,
factor=c(rep("cond1",n1),rep("cond2",n2)),norm=TRUE,depth=depth,
method=filter,cv.cutoff=cv.cutoff,cpm=cpm)
}
else {
datos.filt <- datitos.norm
}
datos1.filt <- datos.filt[,seq_len(n1)]
datos2.filt <- datos.filt[,(n1+1):(n1+n2)]
#-------------------------------------------------------------------------#
## Noise distribution
Zr <- NULL
if (n1+n2 <= 9) { # sharing information within clusters
Zr=share.info(mydata=datos.filt,n1=n1,n2=n2,r=r,nclust=nclust)
}
else { # r permutations
for (i in seq_len(r)) {
print(paste("r =",i))
mipermu <- sample(seq_len(n1+n2))
mipermu <- datos.filt[,mipermu]
mean1 <- rowMeans(mipermu[,seq_len(n1)])
mean2 <- rowMeans(mipermu[,(n1+1):(n1+n2)])
sd1 <- apply(mipermu[,seq_len(n1)],1,sd)
sd2 <- apply(mipermu[,(n1+1):(n1+n2)],1,sd)
myparam <- list("n"=c(n1,n2),"sd"=cbind(sd1,sd2))
MDperm <- MDbio(dat=cbind(mean1,mean2),param=myparam,a0per=a0per)
Zr <- cbind(Zr,myDfunction(mydif=MDperm$D,myrat=MDperm$M,stat=1,
coef=0.5))
}
}
#-------------------------------------------------------------------------#
## Z-score for different experimental conditions (SIGNAL)
mean1 <- rowMeans(as.matrix(datos1.filt))
mean2 <- rowMeans(as.matrix(datos2.filt))
sd1 <- apply(as.matrix(datos1.filt),1,sd)
sd2 <- apply(as.matrix(datos2.filt),1,sd)
myparam=list("n"=c(n1,n2),"sd"=cbind(sd1,sd2))
MDs <- MDbio(dat=cbind(mean1,mean2),param=myparam,a0per=a0per)
Zs <- myDfunction(mydif=MDs$D,myrat=MDs$M,stat=1,coef=0.5)
#-------------------------------------------------------------------------#
## Completing M and D (in signal)
lev1 <- mean1[todos]
lev2 <- mean2[todos]
names(lev1) <- names(lev2) <- todos
Zs <- as.numeric(Zs)
names(Zs) <- rownames(MDs$M)
Zs <- Zs[todos]
names(Zs) <- todos
## Computing Zn
Zn <- as.numeric(Zr)
#-------------------------------------------------------------------------#
## Results
list("k"=k,"comp"=comparison,"Level1"=lev1,"Level2"=lev2,"Zs"=Zs,"Zn"=Zn)
}
## Function to summarize difference and ratio information (D and D0)
myDfunction <- function (mydif,myrat,stat,coef) {
if (stat == 1) { # linear combination of difference and ratio
myDvalues=coef*mydif + (1-coef)*myrat
}
if (stat == 2) { # distance to origin from (ratio,difference)
myDvalues=sign(mydif) * sqrt((mydif)^2 + (myrat)^2)
}
myDvalues
}
################################################################################
################################################################################
# ARSyNcomponents.R
# This program selects the number of components that explain more than the
# Variability% If Variability="average" the number of components will be those
# that explain more than the average variation of the principal components
# For residuals model the number of components selected are
# beta*average-variability.
ARSyNcomponents <- function(asca=asca,Variability=0.75,beta=2) {
MODEL<-asca[-length(asca)]
M<-length(MODEL)-1
output<-NULL
if (Variability=="average") {
for (i in seq_len(M)) {
lim <- 1/rankMatrix(MODEL[[i]]$X)[1]
t <- table(MODEL[[i]]$var.exp[,1]>lim)
if (length(t)==1) {
t[2]=0
}
t <- t[2]
names(t)<-names(MODEL)[i]
output<-c(output,t)
}
}
if (Variability!="average") {
lim<-Variability
for (i in seq_len(M)) {
t <- which(MODEL[[i]]$var.exp[,2]>lim)[1]
names(t) <- names(MODEL)[i]
output <- c(output,t)
}
}
### Residuals model
i <- M+1
lim <- beta*1/rankMatrix(MODEL[[i]]$X)[1]
t <- table(MODEL[[i]]$var.exp[,1]>lim)
if (length(t)==1) {
t[2]=0
}
t <- t[2]
names(t)<-names(MODEL)[i]
output <- c(output,t)
output
}
################################################################################
################################################################################
# ARSyNmodel.R
ARSyNmodel <- function(X=X,Factors=2,Designa=Designa,Designb=Designb,
Designc=Designc,Variability="average",Join=TRUE,Interaction=TRUE,beta=2) {
if (Factors==1) {
Fac0 <- c(1,2)
names(Fac0) <- c("Model.a","Model.res")
asca0 <- ASCA.1f(X=X,Designa=Designa,Fac=Fac0)
Fac <- ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.1f(X=X,Designa=Designa,Fac=Fac0)
}
if (Factors==2) {
Fac0 <- c(1,2,2,2)
names(Fac0) <- c("Model.a","Model.b","Model.ab","Model.res")
if (Join[1]) {
names(Fac0)[3] <- c("Model.bab")
}
asca0 <- ASCA.2f(X=X,Designa=Designa,Designb=Designb,Fac=Fac0,Join=Join,
Interaction=Interaction)
Fac <- ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.2f(X=X,Designa=Designa,Designb=Designb,Fac=Fac0,Join=Join,
Interaction=Interaction)
}
if (Factors==3) {
Fac0 <- c(0,2,2,2,2,2,2,2)
names(Fac0) <- c("Model.a","Model.b","Model.c","Model.ab","Model.ac",
"Model.bc","Model.abc","Model.res")
if (Join[1]) {
names(Fac0)[4]<-c("Model.bab")
}
if (Join[2]) {
names(Fac0)[5]<-c("Model.cac")
}
asca0 <- ASCA.3f(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=Fac0,Join=Join,Interaction=Interaction)
Fac<-ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.3f(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=Fac0,Join=Join,Interaction=Interaction)
}
Input <- asca$Input
asca$Input <- c(Input,Factors)
names(asca$Input) <- c(names(Input),"Factors")
output <- asca
output
}
################################################################################
################################################################################
# ARSyNseq.R
ARSyNseq <- function(data,factor=NULL,batch=FALSE,norm="rpkm",logtransf=FALSE,
Variability=0.75,beta=2) {
Join <- TRUE
Interaction <- TRUE
dat <- as.matrix(assayData(data)$exprs)
long <- featureData(data)@data[rownames(dat),"Length"]
if (is.null(long))
long <- 1000
if (norm == "rpkm") {
dat <- noirpkm(dat,long=long,k=0,lc=1)
}
if (norm == "uqua") {
dat <- noiuqua(dat,long=long,lc=1,k=0)
}
if (norm == "tmm") {
dat <- noitmm(dat,long=long,lc=1,k=0)
}
if (!logtransf) dat <- log(dat + 1)
X <- t(dat) #conditions x genes
#----------------------------------
if (is.null(factor)) {
Covariates <- t(pData(data))
Num.factors <- nrow(Covariates)
labels.factors <- rownames(Covariates)
Design <- list(NULL,NULL,NULL)
for (i in seq_len(Num.factors)) {
x <- as.character(Covariates[i,])
Design[[i]] <- make.ASCA.design(x)
}
}
else {
Covariates <- pData(data)[,factor]
Num.factors <- 1
labels.factors <- factor
Design <- list(NULL,NULL,NULL)
x <- as.character(Covariates)
Design[[1]] <- make.ASCA.design(x)
}
####################################
### --- Execute ASCAmodel
####################################
my.asca <- ARSyNmodel(Factors=Num.factors,X=X,Designa=Design[[1]],
Designb=Design[[2]],Designc=Design[[3]],Join=Join,
Interaction=Interaction,Variability=Variability,beta=beta)
####################################
### --- Writing filtered matrix
####################################
X.filtered <- X
M <- length(my.asca)-1
if (!batch) {
X.filtered <- X.filtered-my.asca[[M]]$TP
}
if(batch) {
X.filtered <- X.filtered-my.asca[[1]]$TP
}
data.filtered <- t(X.filtered)
if (!logtransf)
data.filtered <- exp(data.filtered)+1
exprs(data) <- data.filtered
return(data)
}
################################################################################
# ASCA1f.R
ASCA.1f <- function(X=X,Designa=Designa,Designb=NULL,Designc=NULL,Fac=c(1,2),
Join=NULL,Interaction=NULL) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
Faca <- Fac[1] # number components Model a (time)
Facres <- Fac[2] # number components Residues
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a<-ASCAfun1(Xoff,Designa,Faca)
Xres<-Xoff-Model.a$X
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i], envir=environment())
output <- c(output, list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X,Designa,Designb,Designc,Fac,Join,Interaction)
names(Input)<-c("X", "Designa", "Designb", "Designc", "Fac", "Join",
"Interaction")
Input <- list(Input)
names(Input) <- "Input"
output <- c(output,Input)
output
}
################################################################################
################################################################################
# ASCA2f.R
ASCA.2f <- function(X=X,Designa=Designa,Designb=Designb,Designc=NULL,
Fac=c(1,2,2,2),Join = TRUE,Interaction=TRUE) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
J <- ncol(Designb)
Faca <- Fac[1] # number components Model a (time)
Facb <- Fac[2] # number components Model b (second factor)
Facab <- Fac[3] # number components Model ab (interaction)
Facres <- Fac[4] # number components Residues
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a <- ASCAfun1(Xoff,Designa,Faca)
Xres <- Xoff-Model.a$X
#-------------------------- PART II: Submodel b and ab ---------------------
if (!Join) {
Model.b<-ASCAfun1(Xoff,Designb,Facb)
if (Interaction) {
Model.ab<-ASCAfun2(Xoff,Designa,Designb,Facab)
}
}
if (Join) {
Model.bab <- ASCAfun12(Xoff,Designa,Designb,Facab)
}
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i],envir=environment())
output <- c(output,list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X,Designa,Designb,Designc,Fac,Join,Interaction,Xoff)
names(Input) <- c("X","Designa","Designb","Designc","Fac","Join",
"Interaction","Xoff")
Input <- list(Input)
names(Input) <- "Input"
output <- c(output,Input)
output
}
################################################################################
################################################################################
# ASCA3f.R
ASCA.3f <- function(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=c(1,2,2,2,2,2,2,2),Join=c(TRUE,TRUE),
Interaction=c(TRUE,TRUE,TRUE,TRUE)) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
J <- ncol(Designb)
K <- ncol(Designc)
Faca <- Fac[1]# number components Model a (time)
Facb <- Fac[2] # number components Model b (second factor)
Facc <- Fac[3] # number components Model c (third factor)
Facab <- Fac[4] # number components Model ab (interaction)
Facac <- Fac[5] # number components Model ac (interaction)
Facbc <- Fac[6] # number components Model bc (interaction)
Facabc <- Fac[7] # number components Model abc (interaction)
Facres <- Fac[8]
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a <- ASCAfun1(Xoff,Designa,Faca)
Xres <- Xoff-Model.a$X
#-------------------------- PART II.1: Submodel b.ab------------------------
if(!Join[1]) {
Model.b <- ASCAfun1(Xoff,Designb,Facb)
if (Interaction[1]) {
Model.ab <- ASCAfun2(Xoff,Designa,Designb,Facab)
}
}
if(Join[1]) {
Model.bab<-ASCAfun12(Xoff,Designa,Designb,Facab)
}
#-------------------------- PART II.2: Submodel (c.ac) ---------------------
if(!Join[2]) {
Model.c <- ASCAfun1(Xoff,Designc,Facc)
if (Interaction[2]) {
Model.ac<-ASCAfun2(Xoff,Designa,Designc,Facac)
}
}
if(Join[2]) {
Model.cac<-ASCAfun12(Xoff,Designa,Designc,Facac)
}
#-------------------------- PART II.3: Submodel (bc) -----------------------
if (Interaction[3]) {
Model.bc<-ASCAfun2(Xoff,Designb,Designc,Facbc)
}
#-------------------------- PART II.4: Submodel (abc) ----------------------
if (Interaction[4]) {
Model.abc<-ASCAfun.triple(Xoff,Designa,Designb,Designc,Facabc)
}
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i], envir=environment())
output <- c(output, list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X, Designa, Designb, Designc, Fac, Join,Interaction,Xoff)
names(Input) <- c("X","Designa","Designb","Designc","Fac","Join",
"Interaction","Xoff")
Input <- list(Input)
names(Input) <- "Input"
output<-c(output,Input)
output
}
################################################################################
################################################################################
# ASCAfun1.R
ASCAfun1 <- function(X,Design,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Design) # number of levels in the factor
NK <- NULL
XK <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Design[,i]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i] <- 1
XK[i,] <- sub
}
else {
NK[i]<-nrow(sub)
XK[i,]<-apply(sub,2,mean) }
}
NK<-sqrt(NK)
# Weigh the data of the Submodel with the corresponding number of
# measurement occasions
XKw <- NK*XK
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw<-as.matrix(rep(0,I))
ld<-as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/NK
XKrec <- sc%*%t(ld)
Xa <- NULL
TPa <- NULL
for (i in seq_len(nrow(X))) {
position<-which(Design[i,]==1)
Xa<-rbind(Xa,XK[position,])
TPa<-rbind(TPa,XKrec[position,])
}
Ea <- Xa-TPa
#Leverage & SPE
leverage<-apply(ld^2,1,sum)
SPE<-apply(Ea^2,2,sum)
output <- list(XK,sc,ld,ssq,Xa,TPa,Ea,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfun2.R
ASCAfun2 <- function(X,Desa,Desb,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
XK1[i,] <- sub
}
else {
XK1[i,] <- apply(sub,2,mean)
}
}
XK2 <- matrix(NA,nrow=J,ncol=n)
for (j in seq_len(J)) {
sub <- X[Desb[,j]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
XK2[j,] <- sub
}
else {
XK2[j,]<-apply(sub,2,mean)
}
}
NK <- matrix(NA,nrow=I,ncol=J)
XK <- matrix(NA,nrow=I*J,ncol=n)
k=1
for (j in seq_len(J)){
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i,j] <- 1
XK[k,] <- sub-XK1[i,]-XK2[j,]
}
else {
NK[i,j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)-XK1[i,]-XK2[j,]
}
k <- k+1
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xab <- NULL
TPab <- NULL
for (i in seq_len(nrow(X))) {
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
Xab <- rbind(Xab,XK[I*(position2-1)+position1,])
TPab <- rbind(TPab,XKrec[I*(position2-1)+position1,])
}
Eab <- Xab-TPab
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eab^2,2,sum)
output <- list(XK,sc,ld,ssq,Xab,TPab,Eab,leverage,SPE)
names(output)<-c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfun12.R
ASCAfun12 <- function (X,Desa,Desb,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
XK1[i,] <- apply(sub,2,mean)
}
NK <- matrix(NA,nrow=I,ncol=J)
XK <- matrix(NA,nrow=I*J,ncol=n)
k <- 1
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i,j] <- 1
XK[k,] <- sub-XK1[i,]
}
else {
NK[i,j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)-XK1[i,]
}
k <- k+1
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xab <- NULL
TPab <- NULL
for (i in seq_len(nrow(X))){
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
Xab <- rbind(Xab,XK[I*(position2-1)+position1,])
TPab <- rbind(TPab,XKrec[I*(position2-1)+position1,])
}
Eab <- Xab-TPab
#Leverage & SPE
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eab^2,2,sum)
output <- list(XK,sc,ld,ssq,Xab,TPab,Eab,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfunres.R
ASCAfun.res <- function (X,Fac) {
PCA <- PCA.GENES(X)
sc <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
sc <- as.matrix(sc)
ld <- as.matrix(ld)
}
TPres<-sc%*%t(ld)
if (Fac==0) {
sc <- 0
ld <- 0
TPres<-matrix(0,nrow(X),ncol(X))
}
Eres<-X-TPres
output <- list(sc,ld,ssq,X,TPres,Eres)
names(output) <- c("scores","loadings","var.exp","X","TP","E")
output
}
################################################################################
################################################################################
# ASCAfun-triple.R
ASCAfun.triple <- function(X,Desa,Desb,Desc,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
H <- ncol(Desc) # number of levels in the other factor
#Matrices con medias efectos individuales
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
XK1[i,] <- apply(sub,2,mean)
}
XK2 <- matrix(NA,nrow=J,ncol=n)
for (j in seq_len(J)) {
sub <- X[Desb[,j]==1,]
XK2[j,] <- apply(sub,2,mean)
}
XK3 <- matrix(NA,nrow=H,ncol=n)
for (h in seq_len(H)) {
sub <- X[Desc[,h]==1,]
XK3[h,] <- apply(sub,2,mean)
}
#Matrices con medias de efectos simples
XK12 <- matrix(NA,nrow=I*J,ncol=n)
k <- 1
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
XK12[k,] <- apply(sub,2,mean)
k <- k+1
}
}
XK13 <- matrix(NA,nrow=I*H,ncol=n)
k <- 1
for (h in seq_len(H)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desc[,h])==2,]
XK13[k,] <- apply(sub,2,mean)
k <- k+1
}
}
XK23 <- matrix(NA,nrow=J*H,ncol=n)
k <- 1
for (h in seq_len(H)){
for (j in seq_len(J)){
sub <- X[(Desb[,j]+Desc[,h])==2,]
XK23[k,] <- apply(sub,2,mean)
k <- k+1
}
}
NK <- matrix(NA,nrow=I,ncol=J*H)
XK <- matrix(NA,nrow=I*J*H,ncol=n)
k <- 1
for (h in seq_len(H)) {
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- as.matrix(rbind(X[(Desa[,i]+Desb[,j]+Desc[,h])==3,]))
NK[i,(h-1)*J+j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)+XK1[i,]+
XK2[j,]+XK3[h,]-XK12[(j-1)*I+i,]-XK13[(h-1)*I+i,]-
XK23[(h-1)*J+j,]
k <- k+1
}
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J*H))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xabc <- NULL
TPabc <- NULL
for (i in seq_len(nrow(X))){
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
position3 <- which(Desc[i,]==1)
Xabc <- rbind(Xabc,XK[I*(position2-1)+I*J*(position3-1)+position1,])
TPabc <- rbind(TPabc,XKrec[I*(position2-1)+
I*J*(position3-1)+position1,])
}
Eabc <- Xabc-TPabc
#Leverage & SPE
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eabc^2,2,sum)
output <- list(XK,sc,ld,ssq,Xabc,TPabc,Eabc,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# auxiliar.R
busca <- function (x,S) {
which(S[,1] == x[1] & S[,2] == x[2])
}
int.mult <- function(lista,todos=NULL) {
if (is.null(todos)) {
todos <- unlist(lista)
}
comunes <- todos
for (i in seq_len(length(lista))) {
comunes <- intersect(comunes,lista[[i]])
}
comunes
}
n.menor <- function (x,S1,S2) {
length(which(S1 <= x[1] & S2 <= x[2]))
}
noceros <- function (x,num=TRUE,k=0) {
nn <- length(which(x > k))
if (num) {
nn
}
else {
if(nn > 0) {
which(x > k)
}
else {
NULL
}
}
}
sinceros <- function (datos,k) {
datos <- as.matrix(datos)
datos0 <- as.matrix(datos)
if (is.null(k)) {
mini0 <- min(datos[noceros(datos,num=FALSE,k=0)])
kc <- mini0/2
datos0[datos0 == 0] <- kc
}
else {
datos0[datos0 == 0] <- k
}
datos0
}
sim.samples <- function(counts1,counts2=NULL,pnr=1,nss=5,v=0.02) {
seqdep <- c(sum(counts1),sum(counts2))
num.reads1 <- (pnr + c(-v,v))*seqdep[1]
muestras <- vector("list")
muestras$c1 <- NULL
for (s in seq_len(nss)) {
tama <- round(runif(1,num.reads1[1],num.reads1[2]),0)
muestras$c1 <- cbind(muestras$c1,rmultinom(1,size=tama,prob=counts1))
}
if (!is.null(counts2)) {
num.reads2 <- (pnr + c(-v,v))*seqdep[2]
muestras$c2 <- NULL
for (s in seq_len(nss)) {
tama <- round(runif(1,num.reads2[1],num.reads2[2]),0)
muestras$c2 <- cbind(muestras$c2,
rmultinom(1,size=tama,prob=counts2))
}
}
muestras
}
ranking <- function(results) {
M <- results$M
D <- results$D
prob <- results$prob
## Changing NA by 0
M[is.na(M)] <- 0
D[is.na(D)] <- 0
prob[is.na(prob)] <- 0
## Ranking
ranking5 <- sqrt(M*M + D*D)*sign(M)
## Ranking results
theranking <- data.frame(rownames(results),ranking5)
rownames(theranking) <- NULL
colnames(theranking) <- c("ID","statistic")
theranking
}
plot.y2 <- function(x,yright,yleft,yrightlim=range(yright,na.rm=TRUE),
yleftlim=range(yleft,na.rm=TRUE),xlim=range(x,na.rm=TRUE),xlab=NULL,
yylab=c("",""),lwd=c(2,2),pch=c(1,2),col=c(1,2),type=c("o","o"),
linky=TRUE,smooth=0,bg=c("white","white"),lwds=1,length=10,...,x2=NULL,
yright2=NULL,yleft2=NULL,col2=c(3,4)) {
## Plotting RIGHT axis data
plot(x,yright,axes=FALSE,ylab="",xlab=xlab,ylim=yrightlim,xlim=xlim,
pch=pch[1],type=type[1],lwd=lwd[1],col=col[1],...)
axis(4,pretty(yrightlim,length),col=1,col.axis=1)
if (is.null(yright2) == FALSE) {
points(x2,yright2,type=type[1],pch=pch[1],lwd=lwd[1],col=col2[1],...)
}
if (smooth != 0)
lines(supsmu(x,yright,span=smooth),col=col[1],lwd=lwds,...)
if (yylab[1]=="") {
mtext(deparse(substitute(yright)),side=4,outer=FALSE,line=2,col=1,...)
}
else {
mtext(yylab[1],side=4,outer=FALSE,line=2,col=1,...)
}
par(new=TRUE)
## Plotting LEFT axis data
plot(x,yleft,axes=FALSE,ylab="" ,xlab=xlab,ylim=yleftlim,xlim=xlim,bg=bg[1],
pch=pch[2],type=type[2],lwd=lwd[2],col=col[2],...)
box()
axis(2,pretty(yleftlim,length),col=1,col.axis=1)
if (is.null(yleft2) == FALSE) {
points(x2,yleft2,type=type[2],pch=pch[2],bg=bg[2],lwd=lwd[2],
col=col2[2],...)
}
if (smooth != 0)
lines(supsmu(x,yleft,span=smooth),col=col[2],lwd=lwds,...)
if(yylab[2] == "") {
mtext(deparse(substitute(yleft)),side=2,outer=FALSE,line=2,col=1,...)
}
else {
mtext(yylab[2],side=2,outer=FALSE,line=2,col=1,...)
}
## X-axis
axis(1,at=pretty(xlim,length))
}
logscaling <- function(data,base=2,k=1) {
logmaximo <- round(max(data,na.rm=TRUE),0)
numceros <- nchar(logmaximo)-1
etiquetas <- c(0,10^(seq_len(numceros)))
donde <- log(etiquetas + k,base=base)
data <- log(data + k,base=base)
list("data"=data,"at"=donde,"labels"=etiquetas)
}
miscolores <- colors()[c(554,89,111,512,17,586,132,428,601,568,86,390)]
################################################################################
################################################################################
# biodetection.plot.R
biodetection.dat <- function(input,factor=NULL,k=0) {
if (inherits(input,"eSet") == FALSE)
stop("Error. The input data must be an eSet object\n")
if (any(!is.na(featureData(input)@data$Biotype)) == FALSE)
stop ("No biological classification was provided.\nPlease run ",
"addData() function to add this information\n")
if (!is.null(assayData(input)$exprs)) {
dat <- as.matrix(assayData(input)$exprs)
mysamples <- colnames(assayData(input)$exprs)
}
else {
dat <- as.matrix(assayData(input)$counts)
mysamples <- colnames(assayData(input)$counts)
}
numgenes <- nrow(dat)
if (is.null(factor)) { # per sample
cat("Biotypes detection is to be computed for:\n")
print(colnames(dat))
biotablas <- vector("list",length=NCOL(dat))
names(biotablas) <- colnames(dat)
}
else { # per condition
mifactor <- pData(input)[,factor]
niveles <- levels(mifactor)
cat("Biotypes detection is to be computed for:\n")
print(niveles)
biotablas <- vector("list",length=length(niveles))
names(biotablas) <- niveles
dat <- vapply(niveles,
function (k) rowSums(as.matrix(dat[,grep(k,mifactor)])),numeric(1))
}
infobio <- as.character(featureData(input)@data$Biotype)
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
for (i in seq_len(length(biotablas))) {
detect <- dat[,i] > k
perdet1 <- genome*table(infobio,detect)[names(genome),"TRUE"]/
table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(genome),"TRUE"] /
sum(table(infobio,detect)[,"TRUE"])
biotablas[[i]] <- as.matrix(rbind(perdet1[ordre],perdet2[ordre]))
rownames(biotablas[[i]]) <- c("detectionVSgenome","detectionVSsample")
}
mybiotable <- list("genome"=genome[ordre],"biotables"=biotablas,
"genomesize"=numgenes)
mybiotable
}
biodetection.plot <- function(dat,samples=c(1,2),
plottype=c("persample","comparison"),toplot="protein_coding",
toreport=FALSE,...) {
mypar=par(no.readonly=TRUE)
plottype=match.arg(plottype)
if (length(samples) > 2) {
stop("ERROR: This function cannot generate plots for more than 2 ",
"samples.\nPlease,use it as many times as needed to generate the ",
"plots for all your samples.\n")
}
if (is.numeric(samples))
samples=names(dat$biotables)[samples]
biotable1 <- rbind(dat$genome,dat$biotables[[samples[1]]],
rep(0,length(dat$genome)))
# Computing ylim for left and right axis
if (ncol(biotable1) >= 3) {
ymaxL <- ceiling(max(biotable1[,c(1,2,3)],na.rm=TRUE))
ymaxR <- max(biotable1[,-c(1,2,3)],na.rm=TRUE)
}
else {
ymaxL <- ceiling(max(biotable1,na.rm=TRUE))
ymaxR=0
}
if (length(samples) == 2) {
biotable2 <- rbind(dat$genome,dat$biotables[[samples[2]]],
rep(0,length(dat$genome)))
if (ncol(biotable2) >= 3) {
ymax2 <- ceiling(max(biotable2[,c(1,2,3)],na.rm=TRUE))
ymax2sin <- max(biotable2[,-c(1,2,3)],na.rm=TRUE)
ymaxR <- ceiling(max(ymaxR,ymax2sin))
}
else {
ymax2 <- ceiling(max(biotable2,na.rm=TRUE))
}
ymaxL=max(ymaxL,ymax2)
}
# Rescaling biotables (datos2)
if (length(samples) == 2) {
if (ncol(biotable2) >= 3)
biotable2[,-c(1,2,3)] <- biotable2[,-c(1,2,3)]*ymaxL/ymaxR
}
# Rescaling biotables (datos1)
if (ncol(biotable1) >= 3)
biotable1[,-c(1,2,3)] <- biotable1[,-c(1,2,3)]*ymaxL/ymaxR
## PLOTS
if (length(samples) == 1) { # Plot (1 sample) - 2 scales
par(mar=c(11,4,2,2))
barplot(biotable1[c(1,3),],main=samples[1],xlab=NULL,ylab="%features",
axis.lty=1,legend=FALSE,beside=TRUE,col=c("grey",2),las=2,
ylim=c(0,ymaxL),border=c("grey",2))
barplot(biotable1[c(2,4),],main=samples[1],xlab=NULL,ylab="%features",
axis.lty=1,legend=FALSE,beside=TRUE,col=c(2,1),las=2,density=30,
ylim=c(0,ymaxL),border=2,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("% in genome","detected","% in sample"))
}
else { # Plot (2 samples)
par(mar=c(11,4,2,2))
if (plottype == "persample") {
barplot(biotable1[c(1,3),],main=samples[1],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",2),las=2,ylim=c(0,ymaxL),border=c("grey",2))
barplot(biotable1[c(2,4),],main=samples[1],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c(2,1),las=2,density=30,ylim=c(0,ymaxL),border=2,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("% in genome","detected","% in sample"))
# Datos2
barplot(biotable2[c(1,3),],main=samples[2],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",4),las=2,ylim=c(0,ymaxL),border=c("grey",4))
barplot(biotable2[c(2,4),],main=samples[2],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,col=c(4,1),
las=2,density=30,ylim=c(0,ymaxL),border=4,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("% in genome","detected","% in sample"))
}
# A plot comparing two samples with regard to genome and for % in sample
if (plottype == "comparison") {
lefttable <- rbind(100*dat$biotables[[samples[1]]][1,]/dat$genome,
100*dat$biotables[[samples[2]]][1,]/dat$genome)
righttable <- rbind(dat$biotables[[samples[1]]][2,],
dat$biotables[[samples[2]]][2,])
if (length(toplot) > 1) {
toplot <- toplot[1]
print("WARNING: More than one biotype was provided, the ",
"proportion test will only by applied to the first ",
"biotype.")
}
if ((toplot != 1) && (toplot != "global")) {
numgenes <- dat$genomesize
myx <- round(righttable[,toplot]*numgenes/100,0)
mytest <- prop.test(x=myx,n=rep(numgenes,2),
alternative="two.sided")
if (is.numeric(toplot))
toplot <- colnames(righttable)[toplot]
}
asumar <- colSums(righttable)
asumar <- which(asumar < 0.25)
if (length(asumar) > 1) {
righttable <- cbind(righttable[,-asumar],
rowSums(righttable[,asumar]))
colnames(righttable)[ncol(righttable)] <- "Others"
}
# Detection in the genome
bbb <- barplot(lefttable,main="Biotype detection over genome total",
xlab=NULL,ylab="% detected features",axis.lty=1,legend=FALSE,
cex.names=0.8,beside=TRUE,col=c(2,4),las=2,density=80,
border=c(2,4),ylim=c(0,100))
bbb <- colSums(bbb)/2
lines(bbb,dat$genome,pch=20,type="o",lwd=2)
# %detection in the sample
barplot(righttable,main="Relative biotype abundance in sample",
xlab=NULL,ylab="Relative % biotypes",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(2,4),las=2,border=c(2,4))
legend(x="topright",bty="n",horiz=FALSE,pch=c(15,15,20),
lwd=c(NA,NA,1),legend=c(samples,"% in genome"),col=c(2,4,1))
if ((toplot != 1) && (toplot != "global")) {
print(paste("Percentage of",toplot,"biotype in each sample:"))
names(mytest$estimate) <- samples
print(round(mytest$estimate*100,4))
print(paste("Confidence interval at 95% for the difference of",
"percentages:",samples[1],"-",samples[2]))
print(round(mytest$conf.int[c(1,2)]*100,4))
if (mytest$p.value < 0.05) {
print(paste("The percentage of this biotype is",
"significantly DIFFERENT for these two samples",
"(p-value =",signif(mytest$p.value,4),")."))
}
else {
print(paste("The percentage of this biotype is NOT",
"significantly different for these two samples",
"(p-value =",signif(mytest$p.value,4),")."))
}
}
}
}
# Reset with the default values
if (!toreport)
par(mypar)
}
################################################################################
################################################################################
# cd.plot.R
cd.dat <- function (input,norm=FALSE,refColumn=1) {
if (inherits(input,"eSet") == FALSE)
stop("ERROR: The input data must be an eSet object.\n")
if (!is.null(assayData(input)$exprs)) {
if (ncol( assayData(input)$exprs) < 2)
stop("ERROR: The input object should have at least two samples.\n")
datos <- assayData(input)$exprs
}
else {
if (ncol(assayData(input)$counts) < 2)
stop("ERROR: The input object should have at least two samples.\n")
datos <- assayData(input)$counts
}
ceros = which(rowSums(datos) == 0)
hayceros = (length(ceros) > 0)
if (hayceros) {
print(paste("Warning:",length(ceros),"features with 0 counts in all",
"samples are to be removed for this analysis."))
datos <- datos[-ceros,]
}
## scaling data and/or changing 0 to k
if (norm)
datos <- sinceros(datos, k = NULL)
else
datos <- noirpkm(datos,long=1000,lc=1,k=0.5)
## to plot
data2plot <- log2(datos/datos[,refColumn])
if (is.numeric(refColumn))
refColumn <- colnames(datos)[refColumn]
print(paste("Reference sample is:",refColumn))
MsinRef <- as.matrix(data2plot[,-match(refColumn, colnames(data2plot))])
colnames(MsinRef) <- colnames(data2plot)[-match(refColumn,
colnames(data2plot))]
alpha <- 0.05
alpha <- alpha/ncol(MsinRef)
nperm <- 10^3
bootmed <- vapply(seq_len(nperm),function(k) {
permut <- sample(seq_len(nrow(MsinRef)),replace=TRUE,nrow(MsinRef))
permut <- as.matrix(MsinRef[permut,])
permut <- apply(permut,2,median)
permut
},numeric(ncol(MsinRef)))
if (is.null(dim(bootmed)))
bootmed <- t(as.matrix(bootmed))
bootmed <- t(apply(bootmed,1,quantile,probs=round(c(alpha/2,1-alpha/2),4)))
diagno <- apply(bootmed,1,function (x) {
ddd <- (x[1] <= 0) * (0 <= x[2])
if (ddd == 1)
ddd = "PASSED"
else
ddd = "FAILED"
ddd
})
bootmed <- cbind(bootmed,diagno)
rownames(bootmed) <- colnames(MsinRef)
colnames(bootmed)[3] <- "Diagnostic Test"
print("Confidence intervals for median of M:")
print(bootmed)
if ("FAILED" %in% bootmed[,3])
print(paste0("Diagnostic test: FAILED. Normalization is required ",
"to correct this bias."))
else
print("Diagnostic test: PASSED.")
list("data2plot"=data2plot,"refColumn"=refColumn,"DiagnosticTest"=bootmed)
}
cd.plot <- function (dat,samples=NULL,...) {
refColumn <- dat$refColumn
dat <- dat$data2plot
if (is.null(samples))
samples <- seq_len(ncol(dat))
if (is.numeric(samples))
samples <- colnames(dat)[samples]
samples <- setdiff(samples,refColumn)
if(length(samples) > 12)
stop("Please select 12 samples or less to be plotted (excluding ",
"reference).")
dat <- as.matrix(dat[,samples])
dat.dens <- apply(dat,2,density,adjust=1.5)
limY <- c(0,max(vapply(dat.dens,
function(x) max(x$y,na.rm=TRUE),numeric(1))))
plot(dat.dens[[1]],xlab="M = log2(sample/refsample)",ylab="Density",
lwd=2,ylim=limY,type="l",col=miscolores[1],
main=paste("Reference sample:",refColumn),...)
abline(v=median(dat[,1],na.rm=TRUE),col=miscolores[1],lty=2)
if (length(samples) > 1) {
for (i in 2:length(samples)) {
lines(dat.dens[[i]],col=miscolores[i],lwd=2)
abline(v=median(dat[,i],na.rm=TRUE),col=miscolores[i],lty=i+1)
}
}
legend("topleft",legend=samples,
text.col=miscolores[seq_len(length(samples))],bty="n",lty=1,lwd=2,
col=miscolores[seq_len(length(samples))])
}
################################################################################
################################################################################
# countsbio.plot.R
countsbio.dat <- function (input,biotypes=NULL,factor=NULL,norm=FALSE) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an eSet object\n")
if (!is.null(assayData(input)$exprs))
datos <- assayData(input)$exprs
else
datos <- assayData(input)$counts
depth <- round(colSums(datos)/10^6,1)
names(depth) <- colnames(datos)
ceros <- which(rowSums(datos) == 0)
hayceros <- length(ceros) > 0
if (hayceros) {
warning("Warning:",length(ceros),"features with 0 counts in all ",
"samples are to be removed for this analysis.")
datos0=datos[-ceros,,drop=FALSE]
}
else {
datos0=datos
}
nsam <- NCOL(datos)
if (nsam == 1) {
datos <- as.matrix(datos)
datos0 <- as.matrix(datos0)
}
# Per condition
if (is.null(factor)) { # per sample
print("Count distributions are to be computed for:")
print(colnames(datos))
}
else { # per condition
mifactor <- pData(input)[,factor]
niveles <- levels(mifactor)
print("Counts per million distributions are to be computed for:")
print(niveles)
if (norm) {
datos <- vapply(niveles,function (k) {
rowMeans(as.matrix(datos[,grep(k,mifactor)]))
},numeric(1))
datos0 <- vapply(niveles,function (k) {
rowMeans(as.matrix(datos0[,grep(k,mifactor)]))
},numeric(1))
}
else {
datos <- vapply(niveles,function (k) {
10^6*rowMeans(t(t(datos[,grep(k,mifactor)])/colSums(as.matrix(
datos[,grep(k,mifactor)]))))
},numeric(1))
datos0 <- vapply(niveles,function (k) {
10^6*rowMeans(t(t(datos0[,grep(k,mifactor)])/colSums(as.matrix(
datos0[,grep(k,mifactor)]))))
},numeric(1))
}
colnames(datos) <- colnames(datos0) <- niveles
depth <- vapply(niveles,function (k) {
paste(range(depth[grep(k,mifactor)]),collapse="-")
},character(1))
}
# Biotypes
if (!is.null(featureData(input)$Biotype)) {
if (hayceros) {
infobio0 <- as.character(featureData(input)$Biotype)[-ceros]
}
else {
infobio0 <- as.character(featureData(input)$Biotype)
}
infobio <- as.character(featureData(input)$Biotype)
}
else {
infobio0 <- NULL
infobio <- NULL
}
if (!is.null(infobio)) {
if(is.null(biotypes)) {
biotypes <- unique(infobio)
names(biotypes) <- biotypes
}
# which genes belong to each biotype
biog <- lapply(biotypes,function(x) {
which(is.element(infobio0,x))
})
names(biog) <- biotypes
bionum <- c(NROW(datos0),vapply(biog,length,integer(1)))
names(bionum) <- c("global",names(biotypes))
bio0 <- which(bionum == 0)
if (length(bio0) > 0)
bionum <- bionum[-bio0]
}
else {
biotypes <- NULL
bionum <- NULL
}
# Create the summary matrix information
if (is.null(bionum)) {
resumen <- vector("list",length=1)
names(resumen) <- "global"
}
else {
resumen <- vector("list",length=length(bionum))
names(resumen) <- names(bionum)
}
cuentas <- c(0,1,2,5,10)
if (is.null(factor)) {
if (norm) {
datosCPM <- datos
}
else {
datosCPM=10^6 * t(t(datos)/colSums(as.matrix(datos)))
}
} else { datosCPM=datos }
for (i in seq_len(length(resumen))) {
if (i == 1) {
datosR=datosCPM
} else {
if(!is.null(infobio)) {
datosR <- datosCPM[which(infobio == names(resumen)[i]),]
if (!is(datosR,"matrix")) {
datosR <- t(as.matrix(datosR))
}
}
}
nfeatures <- nrow(datosR)
datosR <- datosR[which(rowSums(datosR) > 0),]
if (!is(datosR,"matrix")) {
datosR <- t(as.matrix(datosR))
}
myglobal <- NULL
mypersample <- NULL
for (kk in seq_len(length(cuentas))) {
mypersample <- rbind(mypersample,apply(datosR,2,function (x) {
length(which(x > cuentas[kk]))
}))
myglobal <- c(myglobal,sum(apply(datosR,1,function (x) {
max(x) > cuentas[kk]
})))
}
mypersample <- round(100*mypersample/nfeatures,1)
mypersample <- rbind(mypersample,depth)
rownames(mypersample) <- seq_len(nrow(mypersample))
myglobal <- c(round(100*myglobal/nfeatures,1),nfeatures)
resumen[[i]] <- data.frame(c(paste("CPM >",cuentas),"depth"),
mypersample,"total"=myglobal)
colnames(resumen[[i]])[1] <- names(resumen)[i]
colnames(resumen[[i]])[2:(ncol(resumen[[i]])-1)] <- colnames(datosR)
}
## results
cosas <- list("result"=datos0,"bionum"=bionum,"biotypes"=infobio0,
"summary"=resumen)
cosas
}
countsbio.plot <- function(dat,samples=c(1,2),toplot="global",
plottype=c("barplot","boxplot"),toreport=FALSE,...) {
# dat: Data coming from countsbio.dat function
# samples: Samples to be plotted. If NULL,all samples are plotted.
# toplot: Name of biotype (including "global") to be plotted.
mypar <- par(no.readonly=TRUE)
plottype <- match.arg(plottype)
## Preparing data
if (is.null(samples)) {
if (NCOL(dat$result) == 1) {
samples=1
} else {
samples <- seq_len(NCOL(dat$result))
}
}
if (is.numeric(toplot))
toplot <- names(dat$summary)[toplot]
if (is.numeric(samples) && !is.null(colnames(dat$result)))
samples <- colnames(dat$result)[samples]
if (plottype == "barplot") {
if ((exists("ylab") && !is.character(ylab)) || !exists("ylab"))
ylab <- ""
datos <- dat$summary[[toplot]]
mytotal <- as.numeric(datos[,"total"])
datos <- as.matrix(datos[,samples])
rownames(datos) <- as.character(dat$summary[[toplot]][,1])
par(mar=c(6,4,4,2))
barplot(as.numeric(datos[1,]),col=miscolores[1],las=2,main="",ylab="",
density=70,ylim=c(0,100),cex.axis=0.8,names.arg="",...)
for (i in 2:(length(mytotal)-2)) {
barplot(as.numeric(datos[i,]),col=miscolores[i],las=2,main="",
ylab="",add=TRUE,density=70,ylim=c(0,100),cex.axis=0.8,
names.arg="",...)
}
bp <- barplot(as.numeric(datos[(length(mytotal)-1),]),
col=miscolores[(length(mytotal)-1)],las=2,
main=paste(toupper(toplot)," (",mytotal[length(mytotal)],")",
sep=""),ylab="Sensitivity (%)",add=TRUE,names.arg=colnames(datos),
cex.axis=0.8,density=70,ylim=c(0,100),cex.names=0.8,...)
for (j in seq_len((length(mytotal)-1)))
abline(h=mytotal[j],col=miscolores[j],lwd=2)
if (length(samples) <= 10) {
mtext(side=3,text=datos["depth",],adj=0.5,at=bp,cex=0.8)
}
else {
mtext(side=3,text=datos["depth",],at=bp,cex=0.7,las=2)
}
legend("top",rownames(datos)[-length(mytotal)],fill=miscolores,
density=70,bty="n",ncol=3)
par(mar=c(5,4,4,4) + 0.1)
}
if (plottype == "boxplot") {
conteos <- as.matrix(dat$result[,samples])
if (is.numeric(samples))
colnames(conteos) <- colnames(dat$result)[samples]
else
colnames(conteos) <- samples
num <- dat$bionum[toplot]
if (is.null(num)) {
if (toplot == "global") {
num=nrow(conteos)
} else {
num=0
}
}
infobio <- dat$biotypes
if (num == 0 && toplot != "global")
stop("Error: No data available. Please,change toplot parameter.")
if ((exists("ylab") && !is.character(ylab)) || !exists("ylab"))
ylab <- "Expression values"
## Plots
if (length(samples) == 1) {
escala <- logscaling(conteos,base=2)
if (is.null(infobio)) {
boxplot(escala$data,col=miscolores[1],ylab=ylab,main="",
yaxt="n",...)
}
else {
par(mar=c(10,4,4,2))
boxplot(escala$data ~ infobio,col=miscolores,ylab=ylab,
main=colnames(conteos),las=2,cex.axis=0.8,cex.lab=0.9,
yaxt="n",...)
cuantos <- dat$bionum[-1]
cuantos <- cuantos[sort(names(cuantos))]
mtext(cuantos,3,at=seq_len(length(cuantos)),cex=0.6,las=2)
}
}
else {
if (toplot != "global")
conteos=conteos[which(infobio == toplot),]
escala <- logscaling(conteos,base=2)
main <- paste(toupper(toplot)," (",num,")",sep="")
par(mar=c(6,4,2,2))
boxplot(escala$data,col=miscolores,ylab=ylab,main=main,las=2,
cex.lab=0.9,cex.axis=0.8,yaxt="n",...)
}
axis(side=2,at=escala$at,labels=escala$labels)
}
if (!toreport)
par(mypar)
}
################################################################################
################################################################################
# dat.R
dat <- function(input,type = c("biodetection","cd","countsbio","GCbias",
"lengthbias","saturation","PCA"),k=0,ndepth=6,factor=NULL,norm=FALSE,
refColumn=1,logtransf = FALSE) {
type <- match.arg(type)
if (type == "biodetection") {
output <- new("Biodetection",
dat=biodetection.dat(input,factor=factor,k=k))
}
if (type == "cd") {
output <- new("CD",dat=cd.dat(input,norm=norm,refColumn=refColumn))
}
if (type == "countsbio") {
output <- new("CountsBio",
dat=countsbio.dat(input,factor=factor,norm=norm))
}
#if (type == "GCbias") {
# output <- new("GCbias",
# dat=GC.dat(input,factor=factor,norm=norm))
#}
#if (type == "lengthbias") {
# output <- new("lengthbias",
# dat=length.dat(input,factor=factor,norm=norm))
#}
if (type == "saturation") {
output <- new("Saturation",dat=saturation.dat(input,k=k,ndepth=ndepth))
}
#if (type == "PCA") {
# output=new("PCA",dat=PCA.dat(input,norm=norm,logtransf=logtransf))
#}
output
}
################################################################################
################################################################################
# DE.plot.R
DE.plot <- function(output,q=NULL,graphic=c("MD","expr","chrom","distr"),
pch=20,cex=0.5,col=1,pch.sel=1,cex.sel=0.6,col.sel=2,log.scale=TRUE,
chromosomes=NULL,join=FALSE,...) {
mypar <- par(no.readonly=TRUE)
if (!is(output,"Output"))
stop("Error. Output argument must contain an object generated by ",
"noiseq or noiseqbio functions.\n")
graphic <- match.arg(graphic)
noiseqbio <- "theta" %in% colnames(output@results[[1]])[c(1,2,3,4)]
## MD plot
if (graphic == "MD") {
if (noiseqbio) {
M <- output@results[[1]][,"log2FC"]
D <- abs(output@results[[1]][,1] - output@results[[1]][,2])+1
names(M) <- names(D) <- rownames(output@results[[1]])
plot(M,D,pch=pch,xlab="M",ylab="D",cex=cex,col=col,log="y",...)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(M[mySelection],D[mySelection],col=col.sel,pch=pch.sel,
cex=cex.sel)
}
}
else {
plot(output@results[[1]][,"M"],(1+output@results[[1]][,"D"]),
pch=pch,xlab="M",ylab="D",cex=cex,col=col,log="y",...)
if (!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(output@results[[1]][mySelection,"M"],
output@results[[1]][mySelection,"D"]+1,col=col.sel,
pch=pch.sel,cex=cex.sel)
}
}
}
## Expression plot: Condition1 vs Condition2
else if (graphic == "expr") {
data <- cbind(output@results[[1]][1],output@results[[1]][2])
rownames(data) <- rownames(output@results[[1]])
colnames(data) <- colnames(output@results[[1]][c(1,2)])
if (log.scale) {
k <- min(data,na.rm=TRUE)
escala <- logscaling(data,base=2,k=k)
plot(escala$data,pch=pch,cex=cex,col=col,yaxt="n",xaxt="n",...)
axis(side=1,at=escala$at,labels=escala$labels)
axis(side=2,at=escala$at,labels=escala$labels)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(escala$data[mySelection,],col=col.sel,pch=pch.sel,
cex=cex.sel)
}
}
else {
plot(data,pch=pch,cex=cex,col=col,...)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(data[mySelection,],col=col.sel,pch=pch.sel,cex=cex.sel)
}
}
}
## MANHATTAN PLOT
else if (graphic == "chrom") {
mydata <- data.frame(as.character(output@results[[1]][,"Chrom"]),
output@results[[1]][,c("GeneStart","GeneEnd")],
output@results[[1]][,c(1,2)])
mydata <- na.omit(mydata)
colnames(mydata) <- c("chr","start","end",colnames(mydata)[-c(1,2,3)])
if (is.null(chromosomes)) { # todos los cromosomas
chromosomes <- unique(mydata$chr)
chromosomes <- sort(chromosomes)
}
print("REMEMBER. You are plotting these chromosomes and in this order:")
print(chromosomes)
# logarithmic scale
if (log.scale) {
if (min(mydata[,-c(1,2,3)],na.rm=TRUE) < 1) {
kk <- -min(mydata[,-c(1,2,3)],na.rm=TRUE)+1
}
else {
kk <- 0
}
mydata[,-c(1,2,3)] <- log2(mydata[,-c(1,2,3)]+kk)
}
# Selecting chromosomes and ordering positions
ordenat <- NULL
for (cromo in chromosomes) {
myselec <- mydata[mydata[,"chr"] == cromo,]
myselec <- myselec[order(myselec[,"start"]),]
ordenat <- rbind(ordenat,myselec)
}
sel.ord <- NULL
if (!is.null(q)) {
# up-regulated in first condition
cond1 <- rownames(degenes(output,q,M="up"))
# up-regulated in second condition
cond2 <- rownames(degenes(output,q,M="down"))
cond1 <- (rownames(ordenat) %in% cond1)*1
cond2 <- (rownames(ordenat) %in% cond2)*1
sel.ord <- cbind(cond1,cond2)
rownames(sel.ord) <- rownames(ordenat)
}
chr.long <- aggregate(ordenat[,"end"],
by=list(as.character(ordenat[,"chr"])),max)
chr.long <- chr.long[match(chromosomes,chr.long[,1]),]
if (join) { # si todos los cromosomas van en el mismo plot
total.long <- sum(as.numeric(chr.long$x))
chr.start <- cumsum(c(1,chr.long$x[-length(chr.long$x)]))
names(chr.start) <- chr.long[,1]
plot(c(1,total.long),c(-max(ordenat[,5]),max(ordenat[,4])),
type="n",xlab="",ylab="Expression data",xaxt="n")
axis(side=1,at=chr.start,labels=chr.long[,1],font=2)
abline(h=0,lty=2,lwd=0.5)
for (ch in chromosomes) {
dat.chr <- ordenat[which(ordenat[,"chr"] == ch),]
dat.chr[,c("start","end")] <- dat.chr[,c("start","end")] +
chr.start[ch] - 1
rect(xleft=dat.chr[,"start"],ybottom=0,xright=dat.chr[,"end"],
ytop=dat.chr[,4],col="grey",border=NA)
rect(xleft=dat.chr[,"start"],ybottom=-dat.chr[,5],
xright=dat.chr[,"end"],ytop=0,col="grey",border=NA)
if (!is.null(q)) {
aux <- which(rownames(sel.ord) %in% rownames(dat.chr))
sel.chr1 <- dat.chr[,4]*sel.ord[aux,1]
sel.chr2 <- -dat.chr[,5]*sel.ord[aux,2]
rect(xleft=dat.chr[,"start"],ybottom=0,
xright=dat.chr[,"end"],ytop=sel.chr1,col=2,border=NA)
rect(xleft=dat.chr[,"start"],ybottom=sel.chr2,
xright=dat.chr[,"end"],ytop=0,col=3,border=NA)
}
segments(x0=dat.chr[,"start"],y0=0,x1=dat.chr[,"end"],y1=0,
col=4,lwd=0.5) # annotated genes
}
text(c(1,1),0.9*c(max(ordenat[,4]),-max(ordenat[,5])),
colnames(mydata)[4:5],font=3,adj=0,col=2:3)
# a plot for each chromosome
}
else {
num.chr <- length(chromosomes)
k <- 20
long.prop <- round((chr.long$x / min(chr.long$x))*k,0)
while (max(long.prop)*num.chr > 500 | max(long.prop) > 50) {
k <- k-1
long.prop <- round((chr.long$x / min(chr.long$x))*k,0)
}
forlayout <- matrix(0,num.chr,max(long.prop))
for (i in seq_len(num.chr)) {
forlayout[i,seq_len(long.prop[i])] <- i
}
layout(forlayout)
if (num.chr > 1)
par(mar=c(2,4.1,0.1,0.1))
miylim <- c(-max(ordenat[,5],na.rm=TRUE),
max(ordenat[,4],na.rm=TRUE))
for (i in seq_len(num.chr)) {
apintar <- ordenat[which(ordenat[,"chr"] == chromosomes[i]),2:5]
plot(c(1,chr.long[i,2]),miylim,type="n",xlab="",
ylab=chromosomes[i],xaxt="n",font.lab=2,cex.lab=1.3)
rect(xleft=apintar[,"start"],ybottom=0,xright=apintar[,"end"],
ytop=apintar[,3],col="grey",border=NA)
rect(xleft=apintar[,"start"],ybottom=-apintar[,4],
xright=apintar[,"end"],ytop=0,col="grey",border=NA)
if (!is.null(q)) {
aux <- which(rownames(sel.ord) %in% rownames(apintar))
asel <- sel.ord[aux,]*apintar[,3:4]
rect(xleft=apintar[,"start"],ybottom=0,
xright=apintar[,"end"],ytop=asel[,1],col=2,border=NA)
rect(xleft=apintar[,"start"],ybottom=-asel[,2],
xright=apintar[,"end"],ytop=0,col=3,border=NA)
}
abline(h=0,lty=2,lwd=0.5)
segments(x0=apintar[,"start"],y0=0,x1=apintar[,"end"],y1=0,
col=4,lwd=0.5) # annotated genes
etiq <- mypretty(c(1,chr.long[i,2]),n=10)
axis(side=1,at=etiq,labels=etiq)
text(c(1,1),0.9*miylim,colnames(mydata)[5:4],font=3,adj=0,
col=3:2)
}
}
}
## DEG distribution across biotypes/chromosomes
else if (graphic == "distr") {
if(!is.null(q)) { # Computing DEG
mySelection <- rownames(degenes(output,q))
detect <- rownames(output@results[[1]]) %in% mySelection
}
else {
stop("You must specify a valid value for q\n")
}
if ("Chrom" %in% colnames(output@results[[1]])) {
numplot <- 1
infobio <- output@results[[1]][,"Chrom"]
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
perdet1 <- genome*table(infobio,detect)[names(
genome),2]/table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(
genome),2]/sum(table(infobio,detect)[,2])
ceros <- rep(0,length(genome))
biotable1 <- as.matrix(rbind(genome[ordre],perdet1[ordre],
perdet2[ordre],ceros))
rownames(biotable1) <- c("genome","degVSgenome","deg","ceros")
if (!is.null(chromosomes))
biotable1=biotable1[,chromosomes]
ymaxL1 <- ceiling(max(biotable1,na.rm=TRUE))
}
else {
numplot <- 0
}
if ("Biotype" %in% colnames(output@results[[1]])) {
numplot <- c(numplot,1)
infobio <- output@results[[1]][,"Biotype"]
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
perdet1 <- genome*table(infobio,detect)[names(
genome),2]/table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(
genome),2]/sum(table(infobio,detect)[,2])
ceros <- rep(0,length(genome))
biotable2 <- as.matrix(rbind(genome[ordre],perdet1[ordre],
perdet2[ordre],ceros))
rownames(biotable2) <- c("genome","degVSgenome","deg","ceros")
higher2 <- which(biotable2[1,] > 2)
lower2 <- which(biotable2[1,] <= 2)
if (length(higher2) > 0) {
ymaxL2 <- ceiling(max(biotable2[,higher2],na.rm=TRUE))
if (length(lower2) > 0) {
ymaxR2 <- ceiling(max(biotable2[,lower2],na.rm=TRUE))
biotable2[,lower2] <- biotable2[,lower2]*ymaxL2/ymaxR2
}
else {
ymaxR2 <- ymaxL2
}
}
else {
ymaxR2 <- ceiling(max(biotable2[,lower2],na.rm=TRUE))
ymaxL2 <- ymaxR2
}
}
else {
numplot <- c(numplot,0)
}
# Plot
if (sum(numplot) == 0)
stop("Biotype or chromosome information is needed for this plot\n")
if (sum(numplot) == 1) { # 1 Plot
par(mar=c(10,4,2,2))
if (numplot[1] == 1) {
barplot(biotable1[c(1,3),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c("grey",2),las=2,ylim=c(0,ymaxL1),
border=c("grey",2))
barplot(biotable1[c(2,4),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(2,1),las=2,density=30,ylim=c(0,ymaxL1),
xborder=2,add=TRUE)
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("%Chrom in genome","%DEG in Chrom",
"%Chrom in DEG"))
}
if (numplot[2] == 1) {
barplot(biotable2[c(1,3),],
main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c("grey",4),las=2,ylim=c(0,ymaxL2),
border=c("grey",4))
barplot(biotable2[c(2,4),],
main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(4,1),las=2,density=30,ylim=c(0,ymaxL2),
border=4,add=TRUE)
axis(side=4,at=pretty(c(0,ymaxL2),n=5),
labels=round(pretty(c(0,ymaxL2),n=5)*ymaxR2/ymaxL2,1))
if (ymaxR2 != ymaxL2) {
abline(v=3*length(higher2) + 0.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("%Biotype in genome","%DEG in Biotype",
"%Biotype in DEG"))
}
}
if (sum(numplot) == 2) { # 2 Plots
par(mar=c(10,4,2,2),mfrow=c(1,2))
# Chromosomes
barplot(biotable1[c(1,3),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",2),las=2,ylim=c(0,ymaxL1),border=c("grey",2))
barplot(biotable1[c(2,4),],
main="DEG distribution across chromosomes",xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,col=c(2,1),
las=2,density=30,ylim=c(0,ymaxL1),border=2,add=TRUE)
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("%Chrom in genome","%DEG in Chrom","%Chrom in DEG"))
# Biotypes
barplot(biotable2[c(1,3),],main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",4),las=2,ylim=c(0,ymaxL2),border=c("grey",4))
barplot(biotable2[c(2,4),],main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c(4,1),las=2,density=30,ylim=c(0,ymaxL2),border=4,add=TRUE)
axis(side=4,at=pretty(c(0,ymaxL2),n=5),
labels=round(pretty(c(0,ymaxL2),n=5)*ymaxR2/ymaxL2,1))
if (ymaxR2 != ymaxL2) {
abline(v=3*length(higher2) + 0.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("%Biotype in genome","%DEG in Biotype",
"%Biotype in DEG"))
}
}
par(mypar)
}
## Auxiliar function
mypretty <- function(x,n=5) {
mywidth <- diff(x)
mybin <- ceiling(mywidth/(n-1))
mylabels0 <- x[1] + mybin*(0:(n-1))
ndig <- nchar(as.character(mylabels0))
mylabels <- mylabels0
k <- 0
for (i in 2:(n-1)) {
mylabels[i] <- (round(mylabels0[i]/10^(ndig[i]-1),k))*10^(ndig[i]-1)
while (mylabels[i] == mylabels[i-1]) {
k <- k+1
mylabels[i] <- (round(mylabels0[i]/10^(ndig[i]-1),k))*10^(ndig[i]-1)
}
}
mylabels
}
################################################################################
################################################################################
# degenes.R
degenes <- function (object,q=0.95,M=NULL) {
if (!is(object,"Output"))
stop("You must give the object returned by the noiseq function\n")
x <- object@results[[1]]
noiseqbio="theta" %in% colnames(x)[seq_len(4)]
if (noiseqbio) {
y <- na.omit(x[c("theta","prob")])
colnames(y)[1]="M"
} else {
y <- na.omit(x[c("M","D","prob")])
}
if (is.null(M)) {
losdeg <- y[y[,"prob"] > q,]
print(paste(dim(losdeg)[1],"differentially expressed features"))
}
else if (M == "up") {
estos <- y[y[,"M"] > 0,]
losdeg <- estos[estos[,"prob"] > q,]
print(paste(dim(losdeg)[1],
"differentially expressed features (up in first condition)"))
}
else if (M == "down") {
estos <- y[y[,"M"] < 0,]
losdeg <- estos[estos[,"prob"] > q,]
print(paste(dim(losdeg)[1],
"differentially expressed features (down in first condition)"))
}
else {
stop("ERROR! Value for parameter M is not valid. Please,choose ",
"between NULL,'up' or 'down'")
}
# Restore the object with the same "results" structure
losdeg <- x[rownames(losdeg),]
losdeg[order(losdeg[,"prob"],decreasing=TRUE),]
}
################################################################################
################################################################################
# fewrepliactes.R
share.info = function (mydata, n1, n2, r, nclust) {
# clustering data by k-means algorithm
# 1.a) Normalized data
gc()
cl <- suppressWarnings(kmeans(mydata,nclust,nstart=25,iter.max=nclust + 30))
cat("...k-means clustering done\n")
cat(paste("Size of",nclust,"clusters:\n"))
print(cl$size)
# Creating pseudo-data
cluster.data <- lapply(seq_len(nclust),function (k) {
mydata[cl$cluster == k,]
})
# Resampling
npermu <- cl$size*r
npermu <- vapply(npermu,function (x) min(x,1000),numeric(1))
cat("Resampling cluster...")
myres=vector("list",length=nclust)
for (i in seq_len(nclust)) {
print(i)
#if (cl$size[i] > 1) { # OPTION 2.A
# OPTION 2.C: small clusters
if (cl$size[i] > 1 && cl$size[i] < 1000) {
#myres[[i]]=t(sapply(seq_len(npermu[i]),function (j) {
# permu=sample(cluster.data[[i]])
# nn1=n1*cl$size[i]
# nn2=n2*cl$size[i]
# mean1=mean(permu[seq_len(nn1)])
# mean2=mean(permu[(nn1+1):(nn1+nn2)])
# sd1=sd(as.numeric(permu[seq_len(nn1)]))
# sd2=sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
# data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
# "M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
# sd2^2/(mean2^2*log(2)^2*nn2)),
# "D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
#}))
myres[[i]]=t(vapply(seq_len(npermu[i]),function (j) {
permu=sample(cluster.data[[i]])
nn1=n1*cl$size[i]
nn2=n2*cl$size[i]
mean1=mean(permu[seq_len(nn1)])
mean2=mean(permu[(nn1+1):(nn1+nn2)])
sd1=sd(as.numeric(permu[seq_len(nn1)]))
sd2=sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
"M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
sd2^2/(mean2^2*log(2)^2*nn2)),
"D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
},vector("list",4)))
}
if (cl$size[i] >= 1000) { # OPTION 2.C & 2.D: big clusters
# Option 2.D: clustering big clusters
cl2 <- kmeans(cluster.data[[i]],nclust,nstart=25,
iter.max=nclust + 20)
cat(paste("Size of",nclust,"subclusters of cluster:",i)); cat("\n")
print(cl2$size)
subcluster.data <- lapply(seq_len(nclust),function (k) {
cluster.data[[i]][cl2$cluster == k,]
})
npermu2 <- cl2$size*r
npermu2 <- vapply(npermu2,function (x) min(x,1000),numeric(1))
## modified to reduce the number of permutations to be done
myres2 <- vector("list",length=nclust)
for (h in seq_len(nclust)) {
if (cl2$size[h] > 1) {
#myres2[[h]] <- t(sapply(seq_len(npermu2[h]),function (j) {
# permu=sample(subcluster.data[[h]])
# nn1 <- n1*cl2$size[h]
# nn2 <- n2*cl2$size[h]
# mean1 <- mean(permu[seq_len(nn1)])
# mean2 <- mean(permu[(nn1+1):(nn1+nn2)])
# sd1 <- sd(as.numeric(permu[seq_len(nn1)]))
# sd2 <- sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
# data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
# "M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
# sd2^2/(mean2^2*log(2)^2*nn2)),
# "D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
#}))
myres2[[h]] <- t(vapply(seq_len(npermu2[h]),function (j) {
permu=sample(subcluster.data[[h]])
nn1 <- n1*cl2$size[h]
nn2 <- n2*cl2$size[h]
mean1 <- mean(permu[seq_len(nn1)])
mean2 <- mean(permu[(nn1+1):(nn1+nn2)])
sd1 <- sd(as.numeric(permu[seq_len(nn1)]))
sd2 <- sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
"M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
sd2^2/(mean2^2*log(2)^2*nn2)),
"D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
},vector("list",4)))
}
}
myres[[i]] <- do.call("rbind",myres2)
}
}
# Computing Zr for noise
cat("Computing Z for noise...\n")
# 4.A) a0: Global for all R*G permutations
myres <- do.call("rbind",myres)
a0.M <- quantile(as.numeric(myres[,"M.sd"]),probs=0.9,na.rm=TRUE)
a0.D <- quantile(as.numeric(myres[,"D.sd"]),probs=0.9,na.rm=TRUE)
M <- as.numeric(myres[,"M"])/(a0.M + as.numeric(myres[,"M.sd"]))
D <- as.numeric(myres[,"D"])/(a0.D + as.numeric(myres[,"D.sd"]))
(M + D)/2
}
################################################################################
# filter.low.counts.R
CV = function(data) {
100*sd(data,na.rm=TRUE)/mean(data,na.rm=TRUE)
}
filtered.data <- function(dataset,factor,norm=TRUE,depth=NULL,method=1,
cv.cutoff=100,cpm=1,p.adj="fdr") {
dataset0=dataset[rowSums(dataset) > 0,]
dataset=dataset0
if ((method == 3) && (norm)) {
if (is.null(depth)) {
stop("ERROR: Sequencing depth for each column in dataset must ",
"be provided.\n")
}
dataset <- t(t(dataset0)/(colSums(dataset0)/depth))
}
if ((method < 3) && (!norm)) {
dataset <- 10^6*t(t(dataset0)/colSums(dataset0))
}
grupos <- unique(factor)
cumple <- NULL
cat("Filtering out low count features...\n")
for (gg in grupos) {
datos <- as.matrix(dataset[,grep(gg,factor)])
if (method == 1) {
if (ncol(datos) == 1) {
cumplecond=(datos > cpm)
}
else {
cumplecond <-
(apply(datos,1,CV) < cv.cutoff)*(rowMeans(datos) > cpm)
cumplecond[which(is.na(cumplecond) == TRUE)] <- 0
}
cumple <- cbind(cumple,cumplecond)
}
if (method == 2) {
if (ncol(datos) == 1)
stop("ERROR: At least 2 replicates per condition are ",
"required to apply this method.")
mytest <- apply(datos,1,function (x) {
suppressWarnings(wilcox.test(x,alternative="greater",
conf.int=FALSE,mu=0))$"p.value"
})
mytest <- p.adjust(mytest,method=p.adj)
cumple <- cbind(cumple,1*(mytest < 0.05))
}
if (method == 3) {
p0=cpm/10^6
mytest <- apply(datos,1,function (x)
suppressWarnings(prop.test(sum(x),n=sum(datos),p=p0,
alternative="greater"))$"p.value")
mytest <- p.adjust(mytest,method=p.adj)
cumple <- cbind(cumple,1*(mytest < 0.05))
}
}
cumple <- which(rowSums(as.matrix(cumple)) >= 1)
cat(paste(length(cumple),"features are to be kept for differential ",
"expression analysis with filtering method",method))
cat("\n")
dataset0[cumple,]
}
################################################################################
################################################################################
# makeASCAdesign.R
make.ASCA.design <- function(x) {
x <- as.factor(x)
levels <- unique(x)
n <- length(x)
p <- length(levels)
Design <- matrix(0,nrow=n,ncol=p)
for (i in seq_len(n)){
for (j in seq_len(p)){
if (x[i]==levels[j]) {
Design[i,j]=1
}
}
}
colnames(Design)<-levels
output<-Design
output
}
################################################################################
################################################################################
# MD.R
MD <- function (dat=dat,selec=c(seq_len(nrow(dat)))) {
pares <- as.matrix(combn(ncol(dat), 2))
if (NCOL(pares) > 30) {
sub30 <- sample(seq_len(NCOL(pares)),size=30,replace=FALSE)
pares <- pares[,sub30]
}
mm <- NULL
dd <- NULL
for (i in seq_len(ncol(pares))) {
a <- dat[selec,pares[1,i]]
b <- dat[selec,pares[2,i]]
mm <- cbind(mm,log(a/b,2))
dd <- cbind(dd,abs(a-b))
}
list("M"=mm,"D"=dd)
}
MDbio <- function (dat=dat,selec= c(seq_len(nrow(dat))),param=NULL,a0per=0.9) {
pares <- as.matrix(combn(ncol(dat), 2))
mm <- NULL
dd <- NULL
for (i in seq_len(ncol(pares))) {
a <- dat[selec,pares[1,i]]
b <- dat[selec,pares[2,i]]
mm <- cbind(mm, log(a/b, 2))
dd <- cbind(dd, (a-b))
}
## Correcting (M,D)
sd.M <- sqrt(param$sd[,1]^2/(dat[,1]^2*log(2)^2*param$n[1]) +
param$sd[,2]^2/(dat[,2]^2*log(2)^2*param$n[2]))
sd.D = sqrt(param$sd[,1]^2/param$n[1] + param$sd[,2]^2/param$n[2])
if(is.null(a0per)) {
a0.M = a0.D = 0
}
else {
if (a0per == "B") {
B <- 100
a0.M <- B*max(sd.M,na.rm=TRUE)
a0.D <- B*max(sd.D,na.rm=TRUE)
}
else {
a0per <- as.numeric(a0per)
a0.M <- quantile(sd.M, probs = a0per, na.rm = TRUE)
a0.D <- quantile(sd.D, probs = a0per, na.rm = TRUE)
}
}
mm <- mm / (a0.M + sd.M)
dd <- dd / (a0.D + sd.D)
# Results
list("M"=mm,"D"=dd)
}
################################################################################
################################################################################
# noiseq.R
noiseq <- function (input,k=0.5,norm=c("rpkm","uqua","tmm","n"),
replicates=c("technical","biological","no"),factor=NULL,conditions=NULL,
pnr=0.2,nss=5,v=0.02,lc=0) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an object generated by the readData ",
"function\n")
if (is.null(factor))
stop("Error. You must specify the factor to know which conditions ",
"you wish to compare. Please,give the argument 'factor'.\n")
replicates <- match.arg(replicates)
if (replicates == "biological")
print("WARNING: Your experiment has biological replicates. You ",
"should consider using NOISeqBIO instead of NOISeq.")
norm <- match.arg(norm)
# (M,D) for signal and noise
print("Computing (M,D) values...")
miMD <- allMD(input,factor,k=k,replicates=replicates,norm=norm,
conditions=conditions,pnr=pnr,nss=nss,v=v,lc=lc)
## Probability of differential expression
print("Computing probability of differential expression...")
prob.concounts <- probdeg(miMD$Ms,miMD$Ds,miMD$Mn,miMD$Dn)$prob
if (!is.null(assayData(input)$exprs))
todos <- rownames(as.matrix(assayData(input)$exprs))
else
todos <- rownames(as.matrix(assayData(input)$counts))
prob <- prob.concounts[todos]
names(prob) <- todos
## Results
resultat <- data.frame("level_1"=miMD$Level1,"level_2"=miMD$Level2,
"M"=miMD$Ms,"D"=miMD$Ds,"prob"=prob)
rownames(resultat) <- todos
# We change the name of the conditions to "name_mean"
colnames(resultat)[1] <-
paste(unlist(strsplit(miMD$comp," "))[1],"mean",sep="_")
colnames(resultat)[2] <-
paste(unlist(strsplit(miMD$comp," "))[3],"mean",sep="_")
resultat <- data.frame(resultat,"ranking"=ranking(resultat)$statistic)
if (!is.null(featureData(input)@data$Length))
resultat <- data.frame(resultat,
"Length"=as.numeric(as.character(featureData(input)@data[todos,
"Length"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$GC))
resultat <- data.frame(resultat,"GC"=as.numeric(as.character(
featureData(input)@data[todos,"GC"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Chromosome))
resultat <- data.frame(resultat,"Chrom"=as.character(
featureData(input)@data$Chromosome),"GeneStart"=as.numeric(
as.character(featureData(input)@data$GeneStart)),
"GeneEnd"=as.numeric(as.character(featureData(input)@data$GeneEnd)),
stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Biotype))
resultat <- data.frame(resultat,"Biotype"=as.character(
featureData(input)@data[todos,"Biotype"]),stringsAsFactors=FALSE)
#resultat[order(resultat[,5],decreasing=TRUE),]
Output(data=list(resultat),method=norm,k=miMD$k,lc=lc,factor=factor,v=v,
nss=nss,pnr=pnr,comparison=miMD$comp,replicates=replicates)
}
################################################################################
################################################################################
# noiseqbio.R
noiseqbio <- function (input,k=0.5,norm=c("rpkm","uqua","tmm","n"),nclust=15,
plot=FALSE,factor=NULL,conditions=NULL,lc=0,r=50,adj=1.5,a0per=0.9,filter=1,
depth=NULL,cv.cutoff=500,cpm=1) {
if (inherits(input,"eSet") == FALSE)
stop("ERROR: You must give an object generated by the readData ",
"function\n")
if (is.null(factor))
stop("ERROR: You must specify the factor to know which conditions ",
"you wish to compare. Please,give the argument 'factor'.\n")
if (min(table(input@phenoData@data[,factor])) < 2)
stop("ERROR: To run NOISeqBIO at least two replicates per condition ",
"are needed. Please, run NOISeq if there are not enough ",
"replicates in your experiment.\n")
norm <- match.arg(norm)
# Z-scores for signal and noise
cat("Computing Z values...\n")
miMD <- allMDbio(input,factor,k=k,norm=norm,conditions=conditions,lc=lc,
r=r,a0per=a0per,nclust=nclust,filter=filter,depth=depth,
cv.cutoff=cv.cutoff,cpm=cpm)
##### Probability of differential expression
cat("Computing probability of differential expression...\n")
## KDE estimators of f0 and f
desde <- min(c(miMD$Zs,miMD$Zn),na.rm=TRUE)
hasta <- max(c(miMD$Zs,miMD$Zn),na.rm=TRUE)
fdens <- density(miMD$Zs,adjust=adj,n=5000,from=desde,to=hasta,na.rm=TRUE)
f <- approxfun(fdens)
f0dens <- density(miMD$Zn,adjust=adj,n=5000,from=desde,to=hasta,na.rm=TRUE)
f0 <- approxfun(f0dens)
if (f0(0)/f(0) < 1)
print("WARNING: f0(0)/f(0) < 1 => FP with Z~0 will be detected.")
f0.f <- f0(miMD$Zs)/f(miMD$Zs)
## f0,f plot
if (plot) {
plot(f0dens,lwd=2,col=4,main=paste("r=",r,"; a0per=",a0per,sep=""),
xlab="theta")
lines(fdens,lwd=2,col=1)
legend("topleft",c("f","f0"),col=c(1,4),lwd=2)
}
## ESTIMATION of p0
p0 <- min(1/f0.f,na.rm=TRUE)
p0=min(p0,1)
cat(paste("p0 =",p0)); cat("\n")
## PROBABILITY of DIFFERENTIAL EXPRESSION
myprob <- 1 - p0*f0.f
if (min(myprob,na.rm=TRUE) < 0) {
print(summary(f0.f))
}
names(myprob) <- names(miMD$Zs)
cat("Probability\n")
print(summary(myprob))
if (!is.null(assayData(input)$exprs))
todos <- rownames(as.matrix(assayData(input)$exprs))
else
todos <- rownames(as.matrix(assayData(input)$counts))
myprob <- myprob[todos]
names(myprob) <- todos
## Results
resultat <- data.frame("level_1"=miMD$Level1,"level_2"=miMD$Level2,
"theta"=miMD$Zs,"prob"=myprob,"log2FC"=log2(miMD$Level1/miMD$Level2))
rownames(resultat) <- todos
colnames(resultat)[1] <-
paste(unlist(strsplit(miMD$comp," "))[1],"mean",sep="_")
colnames(resultat)[2] <-
paste(unlist(strsplit(miMD$comp," "))[3],"mean",sep="_")
# resultat <- data.frame(resultat,"ranking"=ranking(resultat)$statistic)
if (!is.null(featureData(input)@data$Length))
resultat <- data.frame(resultat,"length"=as.numeric(as.character(
featureData(input)@data[todos,"Length"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$GC))
resultat <- data.frame(resultat,"GC"=as.numeric(as.character(
featureData(input)@data[todos,"GC"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Chromosome))
resultat <- data.frame(resultat,"Chrom"=(as.character(
featureData(input)@data$Chromosome)),
"GeneStart"=as.numeric(as.character(featureData(
input)@data$GeneStart)),"GeneEnd"=as.numeric(as.character(
featureData(input)@data$GeneEnd)),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Biotype))
resultat <- data.frame(resultat,"Biotype"=as.character(
featureData(input)@data[todos,"Biotype"]),stringsAsFactors=FALSE)
Output(data=list(resultat),method=norm,k=miMD$k,lc=lc,factor=factor,
comparison=miMD$comp,replicates="biological",v=0,nss=0,pnr=0)
}
################################################################################
################################################################################
# normalization.R
noitmm <- function (datos,long=1000,lc=0,k=0,refColumn=1,logratioTrim=.3,
sumTrim=0.05,doWeighting=TRUE,Acutoff=-1e10) {
if (!is.null(ncol(long))) {
mynames <- long[,1]
long <- long[,2]
names(long) <- mynames
}
L <- (long/1000)^lc
datos <- datos/L
total <- colSums(as.matrix(datos))
datos0 <- sinceros(datos,k)
if (ncol(as.matrix(datos)) > 1) {
fk <- .calcNormFactors(as.matrix(datos),refColumn=refColumn,
method="TMM",logratioTrim=logratioTrim,sumTrim=sumTrim,
doWeighting=doWeighting,Acutoff=Acutoff)
fk <- fk * (total/mean(total))
datos.norm <- t(t(datos0)/fk)
}
else {
datos.norm <- datos0/L
}
na.omit(datos.norm)
}
noirpkm <- function (datos,long=1000,lc=1,k=0) {
if (!is.null(ncol(long))) {
mynames=long[,1]
long=long[,2]
names(long)=mynames
}
total <- colSums(as.matrix(datos))
datos0 <- sinceros(datos,k)
datos.norm <- (t(t(datos0)/total)*10^6)/((long/1000)^lc)
na.omit(datos.norm)
}
noiuqua <- function (datos,long=1000,lc=0,k=0) {
if (!is.null(ncol(long))) {
mynames=long[,1]
long=long[,2]
names(long)=mynames
}
L <- (long/1000)^lc
datos=datos/L
datos0 <- sinceros(datos,k)
if (ncol(as.matrix(datos)) > 1) {
sumatot <- rowSums(datos)
supertot <- sum(sumatot)
counts0 <- which(sumatot == 0)
if (length(counts0) > 0) {
datitos <- datos[-counts0,]
} else {
datitos <- datos
}
q3 <- apply(datitos,2,quantile,probs=0.75)
d <- q3*supertot/sum(q3)
datos.norm <- t(t(datos0)/d)*10^6
} else {
datos.norm <- datos0/L
}
na.omit(datos.norm)
}
## Taken from the edgeR package with minor modifications
.calcNormFactors <- function(object,method=c("TMM","quantile"),refColumn=NULL,
logratioTrim=.3,sumTrim=0.05,doWeighting=TRUE,Acutoff=-1e10,quantile=0.75) {
method <- match.arg(method)
if( is.matrix(object) ) {
if(is.null(refColumn))
refColumn <- 1
data <- object
libsize <- colSums(data)
}
else {
stop(".calcNormFactors() only operates on 'matrix' objects")
}
f <- switch(method,
TMM = apply(data,2,.calcFactorWeighted,ref=data[,refColumn],
logratioTrim=logratioTrim,sumTrim=sumTrim,doWeighting=doWeighting,
Acutoff=Acutoff),quantile=.calcFactorQuantile(data,libsize,
q=quantile))
f <- f/exp(mean(log(f)))
return(f)
}
.calcFactorQuantile <- function(data,lib.size,q=0.75) {
y <- t(t(data)/lib.size)
f <- apply(y,2,function(x) quantile(x,p=q))
f/exp(mean(log(f)))
}
.calcFactorWeighted <- function(obs,ref,logratioTrim=.3,sumTrim=0.05,
doWeighting=TRUE,Acutoff=-1e10) {
if( all(obs==ref) )
return(1)
obs <- as.numeric(obs)
ref <- as.numeric(ref)
nO <- sum(obs)
nR <- sum(ref)
# log ratio of expression,accounting for library size
logR <- log2((obs/nO)/(ref/nR))
absE <- (log2(obs/nO) + log2(ref/nR))/2 # absolute expression
v <- (nO-obs)/nO/obs + (nR-ref)/nR/ref # estimated asymptotic variance
# remove infinite values,cutoff based on A
fin <- is.finite(logR) & is.finite(absE) & (absE > Acutoff)
logR <- logR[fin]
absE <- absE[fin]
v <- v[fin]
# taken from the original mean() function
n <- sum(fin)
loL <- floor(n * logratioTrim) + 1
hiL <- n + 1 - loL
loS <- floor(n * sumTrim) + 1
hiS <- n + 1 - loS
#keep <- (rank(logR) %in% loL:hiL) & (rank(absE) %in% loS:hiS)
# a fix from leonardo ivan almonacid cardenas,since rank() can return
# non-integer values when there are a lot of ties
keep <- (rank(logR)>=loL & rank(logR)<=hiL) & (rank(absE)>=loS
& rank(absE)<=hiS)
if (doWeighting)
2^(sum(logR[keep]/v[keep],na.rm=TRUE) / sum(1/v[keep],na.rm=TRUE))
else
2^(mean(logR[keep],na.rm=TRUE))
}
################################################################################
################################################################################
# PCA.GENES.R
PCA.GENES <- function(X) {
n <- ncol(X)
p <- nrow(X)
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,p,1))%*%rbind(offset))
eigen <- eigen(Xoff%*%t(Xoff)/(p-1))
var <- cbind(eigen$values/sum(eigen$values),
cumsum(eigen$values/sum(eigen$values)))
loadings2 <- eigen$vectors
scores2 <- t(Xoff)%*%loadings2
normas2 <- sqrt(apply(scores2^2,2,sum))
scores1 <- loadings2%*%diag(normas2)
loadings1 <- scores2%*%diag(1/normas2)
output <- list(eigen,var,scores1,loadings1)
names(output) <- c("eigen","var.exp","scores","loadings")
output
}
################################################################################
################################################################################
# probdeg.R
probdeg <- function (Mg,Dg,Mn,Dn,prec=2) {
tot <- length(Mn) # number of points in noise distribution
gens <- names(Mg)
Mruido <- abs(round(Mn,prec))
Druido <- round(Dn,prec)
Mgen <- abs(round(Mg,prec))
Dgen <- round(Dg,prec)
MDgen <- na.omit(cbind(Mgen, Dgen))
MDunic <- unique(MDgen)
Nres <- apply(MDunic,1,n.menor,S1=Mruido,S2=Druido)
lugares <- apply(MDgen,1,busca,S=MDunic)
Nconj <- Nres[lugares]
names(Nconj) <- names(lugares)
laprob <- Nconj / tot
laprob <- laprob[gens]
names(laprob) <- gens
Nconj <- Nconj[gens]
names(Nconj) <- gens
laprob <- list("prob"=laprob,"numDE"=Nconj,"numNOISE"=tot)
laprob
}
################################################################################
################################################################################
# readData.R
readData <- function(data=NULL,factors=NULL,length=NULL,biotype=NULL,
chromosome=NULL,gc=NULL) {
if (is.null(data))
stop("Expression information must be provided to the readData function")
if (is.null(factors))
stop("Condition information must be provided to the readData funcion")
if (!is.null(length) && !is.vector(length) && !is.data.frame(length)
&& !is.matrix(length))
stop( "The length info should be a vector or a data.frame/matrix.")
if (!is.null(gc) && !is.vector(gc) && !is.data.frame(gc) && !is.matrix(gc))
stop( "The GC content info should be a vector or a data.frame/matrix.")
if (!is.null(chromosome) && ncol(chromosome) != 3)
stop( "The chromosome object should be a matrix or data.frame with ",
"3 columns: chromosome,start position and end position.")
if (!is.null(biotype) && !is.vector(biotype) && !is.data.frame(biotype)
&& !is.matrix(biotype))
stop( "The biotype info should be a vector or a data.frame/matrix.")
countData <- as.matrix(data)
rowNames <- rownames(countData)
if (nrow(factors) == ncol(countData)) {
rownames(factors) <- colnames(countData)
}
else {
stop ("Number of rows in factors must be equal to number of ",
"columns in data.\n")
}
pheno <- AnnotatedDataFrame(data=as.data.frame(factors))
input <- ExpressionSet(assayData=countData,phenoData=pheno)
if (!is.null(length))
input <- addData(data=input,length=length)
if (!is.null(gc))
input <- addData(data=input,gc=gc)
if (!is.null(biotype))
input <- addData(data=input,biotype=biotype)
if (!is.null(chromosome))
input <- addData(data=input,chromosome=chromosome)
input
}
addData <- function(data,length=NULL,biotype=NULL,chromosome=NULL,factors=NULL,
gc=NULL) {
if (!inherits(data,"eSet"))
stop("Error. You must give an eSet object.")
if (!is.null(length) && !is.vector(length) && !is.data.frame(length)
&& !is.matrix(length))
stop( "The length info should be a vector or a data.frame/matrix.")
if (!is.null(gc) && !is.vector(gc) && !is.data.frame(gc) && !is.matrix(gc))
stop( "The GC content info should be a vector or a data.frame/matrix.")
if (!is.null(biotype) && !is.vector(biotype) && !is.data.frame(biotype)
&& !is.matrix(biotype))
stop( "The biotype info should be a vector or a data.frame/matrix.")
if (!is.null(chromosome) && ncol(chromosome) != 3)
stop( "The chromosome object should be a matrix or data.frame with ",
"3 columns: chromosome,start position and end position.")
if (!is.null(assayData(data)$exprs))
rowNames <- rownames(assayData(data)$exprs)
else
rowNames <- rownames(assayData(data)$counts)
# If exists length
if (!is.null(length)) {
Length <- rep(NA,length(rowNames))
names(Length) <- rowNames
if (is.vector(length)) {
Length[rowNames] <- as.numeric(as.character(length[rowNames]))
}
else if (is.data.frame(length) || is.matrix(length)) {
if (ncol(length) == 2) {
# We assume that the feature names are in the first column and
# the length in the second
rownames(length) <- length[,1]
Length[rowNames] <- as.numeric(as.character(length[rowNames,2]))
}
else if (ncol(length) == 1) {
# We assume that the length are in the first column and the
# feature names in the rownames
Length[rowNames] <- as.numeric(as.character(
length[rowNames,1]))
}
else {
stop( "The length matrix/data.frame contains more columns ",
"than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,Length)
}
# If exists gc
if (!is.null(gc)) {
GC <- rep(NA,length(rowNames))
names(GC) <- rowNames
if (is.vector(gc)) {
GC[rowNames] <- as.numeric(as.character(gc[rowNames]))
}
else if (is.data.frame(gc) || is.matrix(gc)) {
if (ncol(gc) == 2) {
# We assume that the feature names are in the first column and
# the GC content in the second
rownames(gc) <- gc[,1]
GC[rowNames] <- as.numeric(as.character(gc[rowNames,2]))
}
else if (ncol(gc) == 1) {
# We assume that the GC contents are in the first column
# and the feature names in the rownames
GC[rowNames] <- as.numeric(as.character(gc[rowNames,1]))
}
else {
stop( "The GC matrix/data.frame contains more columns ",
"than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,GC)
}
# If exists biotype
if (!is.null(biotype)) {
Biotype <- rep(NA,length(rowNames))
names(Biotype) <- rowNames
if (is.vector(biotype)) {
Biotype[rowNames] <- as.character(biotype[rowNames])
}
else if (is.data.frame(biotype) || is.matrix(biotype)) {
if (ncol(biotype) == 2) {
# We assume that the feature names are in the first column and
# the biotypes in the second
rownames(biotype) <- biotype[,1]
Biotype[rowNames] <- as.character( biotype[rowNames,2])
}
else if (ncol(biotype) == 1) {
# We assume that the biotypes are in the first column and
# the feature names in the rownames
Biotype[rowNames] <- as.character( biotype[rowNames,1] )
}
else {
stop("The biotype matrix/data.frame contains more ",
"columns than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,Biotype)
featureData(data)@data$Biotype <-
as.character(featureData(data)@data$Biotype)
}
# If exists chromosome
if (!is.null(chromosome)) {
Chromosome <- GeneStart <- GeneEnd <- rep(NA,length(rowNames))
names(Chromosome) <- names(GeneStart) <- names(GeneEnd) <- rowNames
Chromosome[rowNames] <- as.character(chromosome[rowNames,1])
GeneStart[rowNames] <- as.numeric(as.character(chromosome[rowNames,2]))
GeneEnd[rowNames] <- as.numeric(as.character(chromosome[rowNames,3]))
featureData(data)@data <- cbind(featureData(data)@data,Chromosome,
GeneStart,GeneEnd)
}
# If exists new factors
if (!is.null(factors))
phenoData(data)@data <- cbind(phenoData(data)@data,factors)
data
}
################################################################################
################################################################################
# saturation.plot.R
#### SATURATION PLOTS
## Data for saturation plot (with or without biotypes)
saturation.dat <- function (input,k=0,biotypes=NULL,ndepth=6) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an eSet object\n")
if (!is.null(assayData(input)$exprs))
datos <- assayData(input)$exprs
else
datos <- assayData(input)$counts
if (!is.null(featureData(input)$Biotype))
infobio <- featureData(input)$Biotype
else
infobio <- NULL
nsam <- NCOL(datos)
if (!is.null(infobio)) {
if(is.null(biotypes)) {
biotypes <- unique(infobio)
names(biotypes) <- biotypes
}
} else { biotypes=NULL }
satura <- vector("list",length=length(biotypes)+1)
names(satura) <- c("global",names(biotypes))
ndepth1 <- ceiling(ndepth/2)
datos <- round(datos,0)
datos0 <- datos + 0.2
# Random subsamples for each sample
submuestras <- seq.depth <- vector("list",length=nsam)
names(submuestras) <- names(seq.depth) <- colnames(datos)
for (n in seq_len(nsam)) { # simulating subsamples for each sample
total <- sum(datos[,n]) # total counts in sample n
# simulation for each depth and real depth
varias <- vector("list",length=ndepth+1)
for (i in seq_len(ndepth1)) { # simulating depths < real depth
muestra <- rmultinom(10,size=round(total/(ndepth1+1),0),
prob=datos[,n])
if (i == 1)
varias[[i]] <- muestra
else
varias[[i]] <- varias[[i-1]] + muestra
}
varias[[ndepth1+1]] <- as.matrix(datos[,n])
for (i in (ndepth1+2):(ndepth+1)) { # simulating depths < real depth
muestra <- rmultinom(10,size=round(total/(ndepth1+1),0),
prob=datos0[,n])
if (i == ndepth1+2)
varias[[i]] <- matrix(varias[[i-1]],ncol=10,
nrow=nrow(varias[[i-1]])) + muestra
else
varias[[i]] <- varias[[i-1]] + muestra
}
submuestras[[n]] <- varias
seq.depth[[n]] <- c(round(total/(ndepth1+1),0)*seq_len(ndepth1),total,
round(total/(ndepth1+1),0)*((ndepth1+2):(ndepth+1)))
}
# Global saturation
satura[[1]] <- vector("list",length=nsam)
names(satura[[1]]) <- colnames(datos)
for (n in seq_len(nsam)) { # for each sample
satura[[1]][[n]] <- vapply(submuestras[[n]],function(x) {
mean(apply(x,2,noceros,k=k))
},numeric(1))
}
# Per biotypes
if (!is.null(infobio)) { # if biotypes available
biog <- lapply(biotypes,function(x) {
which(is.element(infobio,x))
})
names(biog) <- names(biotypes)
for (j in 2:length(satura)) { # for each biotype
satura[[j]] <- vector("list",length=nsam)
names(satura[[j]]) <- colnames(datos)
for (n in seq_len(nsam)) { # for each sample
conbio <- lapply(submuestras[[n]],function(x) {
as.matrix(x[biog[[j-1]],])
})
satura[[j]][[n]] <- vapply(conbio,function(x) {
mean(apply(x,2,noceros,k=k))
},numeric(1))
}
}
}
else
biog <- NULL
# computing detection increasing per million reads
newdet <- vector("list",length=1+length(biotypes))
names(newdet) <- c("global",names(biotypes))
for (j in seq_len(length(newdet))) {
newdet[[j]] <- vector("list",length=nsam)
names(newdet[[j]]) <- colnames(datos)
for (n in seq_len(nsam)) {
puntos <- data.frame("x"=seq.depth[[n]],"y"=satura[[j]][[n]])
pendi <- NULL
for(i in 2:nrow(puntos))
pendi <- c(pendi,
(puntos$y[i]-puntos$y[i-1])/(puntos$x[i]-puntos$x[i-1]))
newdet[[j]][[n]] <- c(NA,pendi*1000000)
}
}
bionum <- c(NROW(datos),vapply(biog,length,integer(1)))
names(bionum) <- c("global",names(biog))
# Results at real sequencing depth
real <- vector("list",length=length(satura))
names(real) <- names(satura)
realdepth <- vapply(seq.depth,function (x) x[ndepth1+1],numeric(1))/10^6
for (i in seq_len(length(real))) {
real[[i]] <- data.frame("depth"=realdepth,
"detec"=vapply(satura[[i]],function (x) x[ndepth1+1],numeric(1)))
rownames(real[[i]]) <- colnames(datos)
}
# Results
satura <- list("saturation"=satura,"bionum"=bionum,"depth"=seq.depth,
"newdet"=newdet,"real"=real)
satura
}
################################################################################
pmoulos/metaseqR2-local documentation built on May 21, 2024, 3:46 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 addData readData PCA.GENES .calcFactorWeighted .calcFactorQuantile .calcNormFactors make.ASCA.design filtered.data CV mypretty DE.plot dat countsbio.plot biodetection.plot biodetection.dat logscaling plot.y2 ranking sim.samples int.mult ASCAfun.triple ASCAfun2 ASCAfun1 ASCA.3f ASCA.2f ASCA.1f ARSyNseq ARSyNmodel ARSyNcomponents allMDbio
################################################################################
# classes.R
setClass("Biodetection",representation(dat="list"))
setClass("CD",representation(dat="list"))
setClass("CountsBio",representation(dat="list"))
setClass("GCbias",representation(dat="list"))
setClass("lengthbias",representation(dat="list"))
setClass("Saturation",representation(dat="list"))
setClass("PCA",representation(dat="list"))
setGeneric("explo.plot",function(object,...) standardGeneric("explo.plot"))
setMethod("explo.plot","Biodetection",
function(object,samples=c(1,2),plottype=c("persample","comparison"),
toplot="protein_coding",...)
biodetection.plot(object@dat,samples=samples,plottype=plottype,
toplot=toplot,...))
setMethod("explo.plot","CD",function(object,samples=NULL,...)
cd.plot(object@dat,samples=samples,...))
setMethod("explo.plot","CountsBio",
function(object,samples=c(1,2),toplot="global",
plottype=c("barplot","boxplot"),...)
countsbio.plot(object@dat,samples,toplot,plottype,...))
#setMethod("explo.plot","GCbias",
# function(object,samples=NULL,toplot="global",...)
# GC.plot(object@dat,samples=samples,toplot=toplot,...))
#setMethod("explo.plot","lengthbias",
# function(object,samples=NULL,toplot="global",...)
# length.plot(object@dat,samples=samples,toplot=toplot,...))
#setMethod("explo.plot","Saturation",
# function(object,samples=NULL,toplot=1,yleftlim=NULL,yrightlim=NULL,...)
# saturation.plot(object@dat,samples=samples,toplot=toplot,yleftlim=yleftlim,
# yrightlim=yrightlim,...))
#setMethod("explo.plot","PCA",
# function(object,samples=c(1,2),plottype="scores",factor=NULL)
# PCA.plot(object@dat,samples=samples,plottype=plottype,factor=factor))
# Show methods for exploration objects
setMethod("show","Biodetection",function(object) {
cat("\n Reference genome: \n==========\n")
names(dimnames(object@dat$genome))=NULL
print(object@dat$genome)
for (i in seq_len(length(object@dat$biotables))) {
cat("\n",names(object@dat$biotables)[i],"\n==========\n")
print(object@dat$biotables[[i]])
}
})
setMethod("show","CD",function(object) {
cat("\n Confidence intervals for median of M to compare each sample ",
"to reference:\n=======\n")
print(object@dat$DiagnosticTest)
cat("\n Reference sample is:\n=======\n")
print(object@dat$refColumn)
})
setMethod("show","CountsBio",function(object) {
cat("\n Summary: \n============\n")
print(object@dat$summary[[1]])
})
setMethod("show","GCbias",function(object) {
x <- object@dat$RegressionModels
for (i in seq_len(length(x))) {
print(names(x)[i])
print(summary(x[[i]]))
}
})
setMethod("show","lengthbias",function(object) {
x <- object@dat$RegressionModels
for (i in seq_len(length(x))) {
print(names(x)[i])
print(summary(x[[i]]))
}
})
setMethod("show","Saturation",function(object) {
x <- dat2save(object)
cat("\n Number of detected features at each sequencing ",
"depth: \n============\n")
for (i in seq_len(length(x))) {
print(names(x)[i])
print(x[[i]])
}
})
setMethod("show","PCA",function(object) {
x <- object$result$var.exp
x <- round(x*100,4)
colnames(x) <- c("%Var","Cum %Var")
rownames(x) <- paste("PC",seq_len(nrow(x)))
cat("\n Percentage of total variance explained by each component: ",
"\n============\n")
print(x)
})
# Coercion methods for exploration objects
setGeneric("dat2save",function(object) standardGeneric("dat2save"))
setMethod("dat2save","Biodetection",function(object) object@dat)
setMethod("dat2save","CD",function(object) object@dat)
setMethod("dat2save","CountsBio",function(object) object@dat$summary)
setMethod("dat2save","GCbias",function(object) object@dat$data2plot)
setMethod("dat2save","lengthbias",function(object) object@dat$data2plot)
setMethod("dat2save","Saturation",function(object) {
muestras=vector("list",length=length(object@dat$depth))
names(muestras)=names(object@dat$depth)
for (i in seq_len(length(muestras))) {
muestras[[i]] <- object@dat$depth[[i]]
for (j in seq_len(length(object@dat$saturation))) {
muestras[[i]] <- cbind(muestras[[i]],
object@dat$saturation[[j]][[i]])
}
colnames(muestras[[i]]) <- c("depth",names(object@dat$saturation))
}
muestras
})
setMethod("dat2save","PCA",function(object) object@dat$result)
setClass("myInfo",representation(method="character",k="numeric",
lc="numeric",factor="vector",v="numeric",nss="numeric",pnr="numeric",
comparison="vector",replicates="character"))
setClass("Output",representation(results="list"),contains="myInfo")
setValidity("Output",function(object) {
if (!(is.character(object@method))) {
return(paste("Method must be a string"))
} else if (!(is.numeric(object@k))) {
return(paste("k must be numeric"))
} else if (!(is.numeric(object@lc))) {
return(paste("lc must be numeric"))
} else if (!(is.vector(object@factor))) {
return(paste("Factor must be a vector of strings"))
} else if (!(is.numeric(object@v))) {
return(paste("v must be numeric"))
} else if (!(is.numeric(object@nss))) {
return(paste("nss must be numeric"))
} else if (!(is.numeric(object@pnr))) {
return(paste("pnr must be numeric"))
} else if (!(is.vector(object@comparison))) {
return(paste("Comparison must be a vector of strings"))
} else if (!(is.list(object@results))) {
return(paste("Results must be a list of data.frames"))
} else {
return(TRUE)
}
})
Output <- function (data,method,k,lc,factor,v,nss,pnr,comparison,replicates) {
new("Output",results=data,method=method,k=k,lc=lc,factor=factor,v=v,nss=nss,
pnr=pnr,comparison=comparison,replicates=replicates)
}
setMethod("show","Output",function(object) {
if (object@method == "n")
object@method="none"
for (i in seq_len(length(object@results))) {
cat("\nSummary",i,"\n=========\n")
cat("\nYou are comparing",object@comparison[i],"from",object@factor[i],
"\n\n")
print(head(object@results[[i]][order(object@results[[i]][,5],
decreasing=TRUE),]))
}
cat("\nNormalization\n")
cat("\tmethod:",object@method,"\n")
cat("\tk:",object@k,"\n")
cat("\tlc:",object@lc,"\n")
# Simulated samples
if (object@replicates == "no") {
cat("\nYou are working with simulated replicates:\n")
cat("\tpnr:",object@pnr,"\n")
cat("\tnss:",object@nss,"\n")
cat("\tv:",object@v,"\n")
}
# With biological or technical replicates
else {
cat("\nYou are working with",object@replicates,"replicates\n")
}
})
################################################################################
################################################################################
# allMD.R
allMD <- function (input,factor,conditions,k=0.5,replicates,norm="rpkm",pnr=0.2,
nss=5,v=0.02,lc=0) {
# Check if the factor introduced is already defined
# If the factor introduced is defined and has more than 2 conditions,it will
# check if the conditions specified are defined too
condition_fac <- FALSE
condition_lev <- FALSE
datos1 <- datos2 <- matrix()
for (i in colnames(pData(input))) {
if (factor == i) {
condition_fac <- TRUE
if (!is.factor(pData(input)[,i]))
pData(input)[,i]=as.factor(pData(input)[,i])
if (length(levels(pData(input)[,i])) == 2) {
if (!is.null(assayData(input)$exprs)) {
datos1 <-
assayData(input)$exprs[,which(
pData(input)[,i]==levels(pData(input)[,i])[1]),
drop=FALSE]
datos2 <-
assayData(input)$exprs[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2]),
drop=FALSE]
}
else {
datos1 <-
assayData(input)$counts[,which(
pData(input)[,i]==levels(pData(input)[,i])[1]),
drop=FALSE]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i]==levels(pData(input)[,i])[2]),
drop=FALSE]
}
# Define the comparison string
comparison <- paste(levels(pData(input)[,i])[1],
levels(pData(input)[,i])[2],sep=" - ")
condition_lev <- TRUE
}
else {
if (is.null(conditions))
stop("Error. You must specify which conditions you wish ",
"to compare when the factor has two or more ",
"conditions.\n")
if (length(conditions) != 2)
stop("Error. The argument conditions must contain the 2 ",
"conditions you wish to compare.")
l <- conditions %in% pData(input)[,i]
# If they are defined,they will be TRUE
if (l[1] == TRUE && l[2] == TRUE) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[1]),drop=FALSE]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[2]),drop=FALSE]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[1]),drop=FALSE]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[2]),drop=FALSE]
}
# Define the comparison string
comparison <- paste(conditions[1],conditions[2],sep=" - ")
condition_lev <- TRUE
}
}
}
}
if (!condition_fac)
stop("The factor you have written does not correspond with any of the ",
"ones you have defined.")
if (!condition_lev)
stop("The conditions you have written don't exist in the factor ",
"specified.\n")
# Correction to make it work when there are simulated samples
if (replicates == "no")
replicates <- "technical"
n1 <- ncol(as.matrix(datos1))
n2 <- ncol(as.matrix(datos2))
if (norm == "n") { # no normalization
datos1 <- round(datos1,100)
datos2 <- round(datos2,100)
}
if (is.null(k)) {
m1 <- min(datos1[noceros(datos1,num=FALSE)],na.rm=TRUE)
m2 <- min(datos2[noceros(datos2,num=FALSE)],na.rm=TRUE)
mm <- min(m1,m2)
k <- mm/2
}
# Total counts for each gene:
suma1 <- rowSums(as.matrix(datos1))
suma2 <- rowSums(as.matrix(datos2))
# All genes
todos <- rownames(as.matrix(datos1))
# Genes with counts in any condition
concounts <- names(which(suma1+suma2 > 0))
long <- 1000
g.sinL <- NULL
if (!is.null(featureData(input)@data$Length)) {
g.sinL <- names(which(is.na(featureData(input)@data$Length)))
if (any(!is.na(featureData(input)@data$Length)) == TRUE)
long <- featureData(input)@data[concounts,"Length"]
}
if (replicates == "technical") { ### technical replicates
suma1 <- suma1[concounts]
suma2 <- suma2[concounts]
#----------------------------------------------------------------------#
# Normalization of counts for each condition (aggregating replicates)
if (norm == "rpkm") { # RPKM
suma1.norm <- noirpkm(suma1,long=long,k=k,lc=lc)
suma2.norm <- noirpkm(suma2,long=long,k=k,lc=lc)
}
if (norm == "uqua") {
suma.norm <- noiuqua(cbind(suma1,suma2),long=long,lc=lc,k=k)
suma1.norm <- as.matrix(suma.norm[ ,1])
suma2.norm <- as.matrix(suma.norm[ ,2])
}
if (norm == "tmm") {
suma.norm <- noitmm(as.matrix(cbind(suma1,suma2)),long=long,lc=lc,
k=k)
suma1.norm <- as.matrix(suma.norm[ ,1])
suma2.norm <- as.matrix(suma.norm[ ,2])
}
}
#-------------------------------------------------------------------------#
## Noise distribution
if ((n1+n2)>2) { # with real samples
datitos <- cbind(datos1,datos2)
datitos <- datitos[concounts,]
gens.sin0 <- setdiff(concounts,g.sinL)
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
datos1.norm <- datitos.norm[ ,seq_len(n1)]
datos2.norm <- datitos.norm[ ,(n1+1):(n1+n2)]
if (n1 > 1) {
MD1 <- MD(dat=datos1.norm)
}
else {
MD1 <- NULL
}
if (n2 > 1) {
MD2 <- MD(dat=datos2.norm)
}
else {
MD2 <- NULL
}
}
else { # with simulated samples
if (nss == 0) {
nss <- 5
}
datos.sim <- sim.samples(counts1=sinceros(suma1,k=k),
counts2=sinceros(suma2,k=k),pnr=pnr,nss=nss,v=v)
nn <- vapply(datos.sim,ncol,numeric(1))
dat.sim.norm <- vector("list",length=2)
datitos <- cbind(datos.sim[[1]],datos.sim[[2]])
rownames(datitos)=names(suma1)
sumita <- rowSums(datitos)
g.sin0 <- names(which(sumita > 0))
gens.sin0 <- setdiff(g.sin0,g.sinL)
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
dat.sim.norm[[1]] <- datitos.norm[,seq_len(nn[1])]
dat.sim.norm[[2]] <- datitos.norm[,(nn[1]+1):sum(nn)]
MD1 <- MD(dat=dat.sim.norm[[1]])
MD2 <- MD(dat=dat.sim.norm[[2]])
}
Mr <- c(as.numeric(MD1$M),as.numeric(MD2$M))
Dr <- c(as.numeric(MD1$D),as.numeric(MD2$D))
#-------------------------------------------------------------------------#
## M and D for different experimental conditions
if (replicates == "technical" & norm != "n") {
MDs <- MD(dat=cbind(suma1.norm,suma2.norm))
lev1 <- suma1.norm[,1]
lev1 <- lev1[todos]
lev2 <- suma2.norm[,1]
lev2 <- lev2[todos]
}
else {
if ((n1+n1) == 2) {
datos1.norm <- sinceros(as.matrix(datos1)[concounts,],k=k)
datos2.norm <- sinceros(as.matrix(datos2)[concounts,],k=k)
}
resum1.norm <- rowMeans(as.matrix(datos1.norm))
resum2.norm <- rowMeans(as.matrix(datos2.norm))
lev1 <- resum1.norm[todos]
lev2 <- resum2.norm[todos]
MDs <- MD(dat=cbind(resum1.norm,resum2.norm))
}
## Completing M and D
names(lev1) <- names(lev2) <- todos
Ms <- as.numeric(MDs$M)
names(Ms) <- rownames(MDs$M)
Ms <- Ms[todos]
names(Ms) <- todos
Ds <- as.numeric(MDs$D)
names(Ds) <- rownames(MDs$D)
Ds <- Ds[todos]
names(Ds) <- todos
## Results
list("k"=k,"comp"=comparison,"Level1"=lev1,"Level2"=lev2,"Ms"=Ms,"Ds"=Ds,
"Mn"=Mr,"Dn"=Dr)
}
allMDbio <- function(input,factor,conditions,k=0.5,norm="rpkm",lc=1,r=10,
a0per=0.9,nclust=15,filter=1,depth=NULL,cv.cutoff=0,cpm=1) {
# Check if the factor introduced is already defined
# If the factor introduced is defined and has more than 2 conditions,
# it will check if the conditions specified are defined too
condition_fac <- FALSE
condition_lev <- FALSE
datos1 <- datos2 <- matrix()
for (i in colnames(pData(input))) {
if (factor == i) {
condition_fac <- TRUE
if (!is.factor(pData(input)[,i]))
pData(input)[,i] <- as.factor(pData(input)[,i])
if (length(levels(pData(input)[,i])) == 2) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i]==levels(pData(input)[,i])[1])]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2])]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] ==levels(pData(input)[,i])[1])]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] ==levels(pData(input)[,i])[2])]
}
# Define the comparison string
comparison <- paste(levels(pData(input)[,i])[1],
levels(pData(input)[,i])[2],sep=" - ")
condition_lev <- TRUE
if (!((ncol(datos1) > 1) && (ncol(datos2) > 1)))
stop("Error. NOISeqBIO needs at least 2 biological replicates ",
"per condition.\n")
}
else {
if (is.null(conditions))
stop("Error. You must specify which conditions you wish ",
"to compare when the factor has two or more ",
"conditions.\n")
if (length(conditions) != 2)
stop("Error. The argument conditions must contain the 2 ",
"conditions you wish to compare.")
l <- conditions %in% pData(input)[,i]
# If they are defined,they will be TRUE
if (l[1] == TRUE && l[2] == TRUE) {
if (!is.null(assayData(input)$exprs)) {
datos1 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[1])]
datos2 <- assayData(input)$exprs[,which(
pData(input)[,i] == conditions[2])]
}
else {
datos1 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[1])]
datos2 <- assayData(input)$counts[,which(
pData(input)[,i] == conditions[2])]
}
# Define the comparison string
comparison <- paste(conditions[1],conditions[2],sep=" - ")
condition_lev <- TRUE
}
}
}
}
if (condition_fac == FALSE)
stop("The factor specified does not correspond with any of the ",
"ones you have defined.")
if (condition_lev == FALSE)
stop("The conditions specified don't exist for the factor specified.\n")
#--------------------------------------------------------------------------#
# Number of observations within each condition
n1 <- ncol(as.matrix(datos1))
n2 <- ncol(as.matrix(datos2))
if (max(n1,n2) == 1)
stop("There is only one replicate per condition. Please,use ",
"NOISeq instead of NOISeqBIO.\n")
# Rounding off data
if (norm == "n") { # no normalization
datos1 <- round(datos1,10)
datos2 <- round(datos2,10)
}
# Computing k
if (is.null(k)) {
m1 <- min(datos1[noceros(datos1,num=FALSE)],na.rm=TRUE)
m2 <- min(datos2[noceros(datos2,num=FALSE)],na.rm=TRUE)
k <- min(m1,m2)/2
}
# Total counts for each gene:
suma1 <- rowSums(as.matrix(datos1))
suma2 <- rowSums(as.matrix(datos2))
# Genes with counts in any condition
concounts <- names(which(suma1+suma2 > 0))
# All genes
todos <- rownames(as.matrix(datos1))
# Gene length
long <- 1000
g.sinL <- NULL # genes with no length defined
if (!is.null(featureData(input)@data$Length)) {
g.sinL <- names(which(is.na(featureData(input)@data$Length)))
if (any(!is.na(featureData(input)@data$Length)) == TRUE)
long <- featureData(input)@data[concounts,"Length"]
}
# Genes with counts and with length
gens.sin0 <- setdiff(concounts,g.sinL)
# cond1 and cond2 in the same matrix
datitos <- cbind(datos1,datos2)
datitos <- datitos[concounts,] # selecting only genes with counts
# Sequencing depth when filtering method=3
if (filter == 3 && is.null(depth))
depth <- colSums(datitos)
#-------------------------------------------------------------------------#
#-------------------------------------------------------------------------#
## Normalization
if (norm == "n") { # no normalization
datitos.0 <- sinceros(datitos,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "rpkm") { # RPKM
datitos.0 <- noirpkm(datitos,long=long,k=k,lc=lc)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "uqua") { # Upper Quartile
datitos.0 <- noiuqua(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
if (norm == "tmm") {
datitos.0 <- noitmm(datitos,long=long,lc=lc,k=k)
datitos.norm <- datitos.0[gens.sin0,]
}
#-------------------------------------------------------------------------#
## Filtering out low count features
if (filter != 0) {
datos.filt <- filtered.data(dataset=datitos.norm,
factor=c(rep("cond1",n1),rep("cond2",n2)),norm=TRUE,depth=depth,
method=filter,cv.cutoff=cv.cutoff,cpm=cpm)
}
else {
datos.filt <- datitos.norm
}
datos1.filt <- datos.filt[,seq_len(n1)]
datos2.filt <- datos.filt[,(n1+1):(n1+n2)]
#-------------------------------------------------------------------------#
## Noise distribution
Zr <- NULL
if (n1+n2 <= 9) { # sharing information within clusters
Zr=share.info(mydata=datos.filt,n1=n1,n2=n2,r=r,nclust=nclust)
}
else { # r permutations
for (i in seq_len(r)) {
print(paste("r =",i))
mipermu <- sample(seq_len(n1+n2))
mipermu <- datos.filt[,mipermu]
mean1 <- rowMeans(mipermu[,seq_len(n1)])
mean2 <- rowMeans(mipermu[,(n1+1):(n1+n2)])
sd1 <- apply(mipermu[,seq_len(n1)],1,sd)
sd2 <- apply(mipermu[,(n1+1):(n1+n2)],1,sd)
myparam <- list("n"=c(n1,n2),"sd"=cbind(sd1,sd2))
MDperm <- MDbio(dat=cbind(mean1,mean2),param=myparam,a0per=a0per)
Zr <- cbind(Zr,myDfunction(mydif=MDperm$D,myrat=MDperm$M,stat=1,
coef=0.5))
}
}
#-------------------------------------------------------------------------#
## Z-score for different experimental conditions (SIGNAL)
mean1 <- rowMeans(as.matrix(datos1.filt))
mean2 <- rowMeans(as.matrix(datos2.filt))
sd1 <- apply(as.matrix(datos1.filt),1,sd)
sd2 <- apply(as.matrix(datos2.filt),1,sd)
myparam=list("n"=c(n1,n2),"sd"=cbind(sd1,sd2))
MDs <- MDbio(dat=cbind(mean1,mean2),param=myparam,a0per=a0per)
Zs <- myDfunction(mydif=MDs$D,myrat=MDs$M,stat=1,coef=0.5)
#-------------------------------------------------------------------------#
## Completing M and D (in signal)
lev1 <- mean1[todos]
lev2 <- mean2[todos]
names(lev1) <- names(lev2) <- todos
Zs <- as.numeric(Zs)
names(Zs) <- rownames(MDs$M)
Zs <- Zs[todos]
names(Zs) <- todos
## Computing Zn
Zn <- as.numeric(Zr)
#-------------------------------------------------------------------------#
## Results
list("k"=k,"comp"=comparison,"Level1"=lev1,"Level2"=lev2,"Zs"=Zs,"Zn"=Zn)
}
## Function to summarize difference and ratio information (D and D0)
myDfunction <- function (mydif,myrat,stat,coef) {
if (stat == 1) { # linear combination of difference and ratio
myDvalues=coef*mydif + (1-coef)*myrat
}
if (stat == 2) { # distance to origin from (ratio,difference)
myDvalues=sign(mydif) * sqrt((mydif)^2 + (myrat)^2)
}
myDvalues
}
################################################################################
################################################################################
# ARSyNcomponents.R
# This program selects the number of components that explain more than the
# Variability% If Variability="average" the number of components will be those
# that explain more than the average variation of the principal components
# For residuals model the number of components selected are
# beta*average-variability.
ARSyNcomponents <- function(asca=asca,Variability=0.75,beta=2) {
MODEL<-asca[-length(asca)]
M<-length(MODEL)-1
output<-NULL
if (Variability=="average") {
for (i in seq_len(M)) {
lim <- 1/rankMatrix(MODEL[[i]]$X)[1]
t <- table(MODEL[[i]]$var.exp[,1]>lim)
if (length(t)==1) {
t[2]=0
}
t <- t[2]
names(t)<-names(MODEL)[i]
output<-c(output,t)
}
}
if (Variability!="average") {
lim<-Variability
for (i in seq_len(M)) {
t <- which(MODEL[[i]]$var.exp[,2]>lim)[1]
names(t) <- names(MODEL)[i]
output <- c(output,t)
}
}
### Residuals model
i <- M+1
lim <- beta*1/rankMatrix(MODEL[[i]]$X)[1]
t <- table(MODEL[[i]]$var.exp[,1]>lim)
if (length(t)==1) {
t[2]=0
}
t <- t[2]
names(t)<-names(MODEL)[i]
output <- c(output,t)
output
}
################################################################################
################################################################################
# ARSyNmodel.R
ARSyNmodel <- function(X=X,Factors=2,Designa=Designa,Designb=Designb,
Designc=Designc,Variability="average",Join=TRUE,Interaction=TRUE,beta=2) {
if (Factors==1) {
Fac0 <- c(1,2)
names(Fac0) <- c("Model.a","Model.res")
asca0 <- ASCA.1f(X=X,Designa=Designa,Fac=Fac0)
Fac <- ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.1f(X=X,Designa=Designa,Fac=Fac0)
}
if (Factors==2) {
Fac0 <- c(1,2,2,2)
names(Fac0) <- c("Model.a","Model.b","Model.ab","Model.res")
if (Join[1]) {
names(Fac0)[3] <- c("Model.bab")
}
asca0 <- ASCA.2f(X=X,Designa=Designa,Designb=Designb,Fac=Fac0,Join=Join,
Interaction=Interaction)
Fac <- ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.2f(X=X,Designa=Designa,Designb=Designb,Fac=Fac0,Join=Join,
Interaction=Interaction)
}
if (Factors==3) {
Fac0 <- c(0,2,2,2,2,2,2,2)
names(Fac0) <- c("Model.a","Model.b","Model.c","Model.ab","Model.ac",
"Model.bc","Model.abc","Model.res")
if (Join[1]) {
names(Fac0)[4]<-c("Model.bab")
}
if (Join[2]) {
names(Fac0)[5]<-c("Model.cac")
}
asca0 <- ASCA.3f(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=Fac0,Join=Join,Interaction=Interaction)
Fac<-ARSyNcomponents(asca0,Variability=Variability,beta=beta)
for (i in seq_len(length(Fac))) {
Fac0[names(Fac[i])]<-Fac[names(Fac[i])]
}
asca <- ASCA.3f(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=Fac0,Join=Join,Interaction=Interaction)
}
Input <- asca$Input
asca$Input <- c(Input,Factors)
names(asca$Input) <- c(names(Input),"Factors")
output <- asca
output
}
################################################################################
################################################################################
# ARSyNseq.R
ARSyNseq <- function(data,factor=NULL,batch=FALSE,norm="rpkm",logtransf=FALSE,
Variability=0.75,beta=2) {
Join <- TRUE
Interaction <- TRUE
dat <- as.matrix(assayData(data)$exprs)
long <- featureData(data)@data[rownames(dat),"Length"]
if (is.null(long))
long <- 1000
if (norm == "rpkm") {
dat <- noirpkm(dat,long=long,k=0,lc=1)
}
if (norm == "uqua") {
dat <- noiuqua(dat,long=long,lc=1,k=0)
}
if (norm == "tmm") {
dat <- noitmm(dat,long=long,lc=1,k=0)
}
if (!logtransf) dat <- log(dat + 1)
X <- t(dat) #conditions x genes
#----------------------------------
if (is.null(factor)) {
Covariates <- t(pData(data))
Num.factors <- nrow(Covariates)
labels.factors <- rownames(Covariates)
Design <- list(NULL,NULL,NULL)
for (i in seq_len(Num.factors)) {
x <- as.character(Covariates[i,])
Design[[i]] <- make.ASCA.design(x)
}
}
else {
Covariates <- pData(data)[,factor]
Num.factors <- 1
labels.factors <- factor
Design <- list(NULL,NULL,NULL)
x <- as.character(Covariates)
Design[[1]] <- make.ASCA.design(x)
}
####################################
### --- Execute ASCAmodel
####################################
my.asca <- ARSyNmodel(Factors=Num.factors,X=X,Designa=Design[[1]],
Designb=Design[[2]],Designc=Design[[3]],Join=Join,
Interaction=Interaction,Variability=Variability,beta=beta)
####################################
### --- Writing filtered matrix
####################################
X.filtered <- X
M <- length(my.asca)-1
if (!batch) {
X.filtered <- X.filtered-my.asca[[M]]$TP
}
if(batch) {
X.filtered <- X.filtered-my.asca[[1]]$TP
}
data.filtered <- t(X.filtered)
if (!logtransf)
data.filtered <- exp(data.filtered)+1
exprs(data) <- data.filtered
return(data)
}
################################################################################
# ASCA1f.R
ASCA.1f <- function(X=X,Designa=Designa,Designb=NULL,Designc=NULL,Fac=c(1,2),
Join=NULL,Interaction=NULL) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
Faca <- Fac[1] # number components Model a (time)
Facres <- Fac[2] # number components Residues
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a<-ASCAfun1(Xoff,Designa,Faca)
Xres<-Xoff-Model.a$X
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i], envir=environment())
output <- c(output, list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X,Designa,Designb,Designc,Fac,Join,Interaction)
names(Input)<-c("X", "Designa", "Designb", "Designc", "Fac", "Join",
"Interaction")
Input <- list(Input)
names(Input) <- "Input"
output <- c(output,Input)
output
}
################################################################################
################################################################################
# ASCA2f.R
ASCA.2f <- function(X=X,Designa=Designa,Designb=Designb,Designc=NULL,
Fac=c(1,2,2,2),Join = TRUE,Interaction=TRUE) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
J <- ncol(Designb)
Faca <- Fac[1] # number components Model a (time)
Facb <- Fac[2] # number components Model b (second factor)
Facab <- Fac[3] # number components Model ab (interaction)
Facres <- Fac[4] # number components Residues
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a <- ASCAfun1(Xoff,Designa,Faca)
Xres <- Xoff-Model.a$X
#-------------------------- PART II: Submodel b and ab ---------------------
if (!Join) {
Model.b<-ASCAfun1(Xoff,Designb,Facb)
if (Interaction) {
Model.ab<-ASCAfun2(Xoff,Designa,Designb,Facab)
}
}
if (Join) {
Model.bab <- ASCAfun12(Xoff,Designa,Designb,Facab)
}
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i],envir=environment())
output <- c(output,list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X,Designa,Designb,Designc,Fac,Join,Interaction,Xoff)
names(Input) <- c("X","Designa","Designb","Designc","Fac","Join",
"Interaction","Xoff")
Input <- list(Input)
names(Input) <- "Input"
output <- c(output,Input)
output
}
################################################################################
################################################################################
# ASCA3f.R
ASCA.3f <- function(X=X,Designa=Designa,Designb=Designb,Designc=Designc,
Fac=c(1,2,2,2,2,2,2,2),Join=c(TRUE,TRUE),
Interaction=c(TRUE,TRUE,TRUE,TRUE)) {
n <- ncol(X)
p <- nrow(X)
I <- ncol(Designa)
J <- ncol(Designb)
K <- ncol(Designc)
Faca <- Fac[1]# number components Model a (time)
Facb <- Fac[2] # number components Model b (second factor)
Facc <- Fac[3] # number components Model c (third factor)
Facab <- Fac[4] # number components Model ab (interaction)
Facac <- Fac[5] # number components Model ac (interaction)
Facbc <- Fac[6] # number components Model bc (interaction)
Facabc <- Fac[7] # number components Model abc (interaction)
Facres <- Fac[8]
#----------------------- Calculate Overall Mean ----------------------------
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,nrow=p,ncol=1))%*%rbind(offset))
#----------------------- PART I: Submodel a (TIME) ------------------------
Model.a <- ASCAfun1(Xoff,Designa,Faca)
Xres <- Xoff-Model.a$X
#-------------------------- PART II.1: Submodel b.ab------------------------
if(!Join[1]) {
Model.b <- ASCAfun1(Xoff,Designb,Facb)
if (Interaction[1]) {
Model.ab <- ASCAfun2(Xoff,Designa,Designb,Facab)
}
}
if(Join[1]) {
Model.bab<-ASCAfun12(Xoff,Designa,Designb,Facab)
}
#-------------------------- PART II.2: Submodel (c.ac) ---------------------
if(!Join[2]) {
Model.c <- ASCAfun1(Xoff,Designc,Facc)
if (Interaction[2]) {
Model.ac<-ASCAfun2(Xoff,Designa,Designc,Facac)
}
}
if(Join[2]) {
Model.cac<-ASCAfun12(Xoff,Designa,Designc,Facac)
}
#-------------------------- PART II.3: Submodel (bc) -----------------------
if (Interaction[3]) {
Model.bc<-ASCAfun2(Xoff,Designb,Designc,Facbc)
}
#-------------------------- PART II.4: Submodel (abc) ----------------------
if (Interaction[4]) {
Model.abc<-ASCAfun.triple(Xoff,Designa,Designb,Designc,Facabc)
}
#------------------------Collecting models ---------------------------------
models <- ls(pattern="Model")
output <- vector(mode="list")
Xres <- Xoff
for (i in seq_len(length(models))) {
mymodel <- get(models[i], envir=environment())
output <- c(output, list(mymodel))
Xres <- Xres - mymodel$X
rm(mymodel)
gc()
}
names(output) <- models
#------------------------- PART III: Submodel res --------------------------
Model.res <- ASCAfun.res(Xres,Facres)
Model.res <- list(Model.res)
names(Model.res) <- c("Model.res")
output <- c(output,Model.res)
#------------------------- Add Input data to the Output --------------------
Input <- list(X, Designa, Designb, Designc, Fac, Join,Interaction,Xoff)
names(Input) <- c("X","Designa","Designb","Designc","Fac","Join",
"Interaction","Xoff")
Input <- list(Input)
names(Input) <- "Input"
output<-c(output,Input)
output
}
################################################################################
################################################################################
# ASCAfun1.R
ASCAfun1 <- function(X,Design,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Design) # number of levels in the factor
NK <- NULL
XK <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Design[,i]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i] <- 1
XK[i,] <- sub
}
else {
NK[i]<-nrow(sub)
XK[i,]<-apply(sub,2,mean) }
}
NK<-sqrt(NK)
# Weigh the data of the Submodel with the corresponding number of
# measurement occasions
XKw <- NK*XK
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw<-as.matrix(rep(0,I))
ld<-as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/NK
XKrec <- sc%*%t(ld)
Xa <- NULL
TPa <- NULL
for (i in seq_len(nrow(X))) {
position<-which(Design[i,]==1)
Xa<-rbind(Xa,XK[position,])
TPa<-rbind(TPa,XKrec[position,])
}
Ea <- Xa-TPa
#Leverage & SPE
leverage<-apply(ld^2,1,sum)
SPE<-apply(Ea^2,2,sum)
output <- list(XK,sc,ld,ssq,Xa,TPa,Ea,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfun2.R
ASCAfun2 <- function(X,Desa,Desb,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
XK1[i,] <- sub
}
else {
XK1[i,] <- apply(sub,2,mean)
}
}
XK2 <- matrix(NA,nrow=J,ncol=n)
for (j in seq_len(J)) {
sub <- X[Desb[,j]==1,]
if (is.null(nrow(sub))) { #when there isn't replicates
XK2[j,] <- sub
}
else {
XK2[j,]<-apply(sub,2,mean)
}
}
NK <- matrix(NA,nrow=I,ncol=J)
XK <- matrix(NA,nrow=I*J,ncol=n)
k=1
for (j in seq_len(J)){
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i,j] <- 1
XK[k,] <- sub-XK1[i,]-XK2[j,]
}
else {
NK[i,j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)-XK1[i,]-XK2[j,]
}
k <- k+1
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xab <- NULL
TPab <- NULL
for (i in seq_len(nrow(X))) {
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
Xab <- rbind(Xab,XK[I*(position2-1)+position1,])
TPab <- rbind(TPab,XKrec[I*(position2-1)+position1,])
}
Eab <- Xab-TPab
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eab^2,2,sum)
output <- list(XK,sc,ld,ssq,Xab,TPab,Eab,leverage,SPE)
names(output)<-c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfun12.R
ASCAfun12 <- function (X,Desa,Desb,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
XK1[i,] <- apply(sub,2,mean)
}
NK <- matrix(NA,nrow=I,ncol=J)
XK <- matrix(NA,nrow=I*J,ncol=n)
k <- 1
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
if (is.null(nrow(sub))) { #when there isn't replicates
NK[i,j] <- 1
XK[k,] <- sub-XK1[i,]
}
else {
NK[i,j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)-XK1[i,]
}
k <- k+1
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xab <- NULL
TPab <- NULL
for (i in seq_len(nrow(X))){
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
Xab <- rbind(Xab,XK[I*(position2-1)+position1,])
TPab <- rbind(TPab,XKrec[I*(position2-1)+position1,])
}
Eab <- Xab-TPab
#Leverage & SPE
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eab^2,2,sum)
output <- list(XK,sc,ld,ssq,Xab,TPab,Eab,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# ASCAfunres.R
ASCAfun.res <- function (X,Fac) {
PCA <- PCA.GENES(X)
sc <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
sc <- as.matrix(sc)
ld <- as.matrix(ld)
}
TPres<-sc%*%t(ld)
if (Fac==0) {
sc <- 0
ld <- 0
TPres<-matrix(0,nrow(X),ncol(X))
}
Eres<-X-TPres
output <- list(sc,ld,ssq,X,TPres,Eres)
names(output) <- c("scores","loadings","var.exp","X","TP","E")
output
}
################################################################################
################################################################################
# ASCAfun-triple.R
ASCAfun.triple <- function(X,Desa,Desb,Desc,Fac) {
n <- ncol(X) # number of genes
I <- ncol(Desa) # number of levels in the factor TIME
J <- ncol(Desb) # number of levels in the other factor
H <- ncol(Desc) # number of levels in the other factor
#Matrices con medias efectos individuales
XK1 <- matrix(NA,nrow=I,ncol=n)
for (i in seq_len(I)) {
sub <- X[Desa[,i]==1,]
XK1[i,] <- apply(sub,2,mean)
}
XK2 <- matrix(NA,nrow=J,ncol=n)
for (j in seq_len(J)) {
sub <- X[Desb[,j]==1,]
XK2[j,] <- apply(sub,2,mean)
}
XK3 <- matrix(NA,nrow=H,ncol=n)
for (h in seq_len(H)) {
sub <- X[Desc[,h]==1,]
XK3[h,] <- apply(sub,2,mean)
}
#Matrices con medias de efectos simples
XK12 <- matrix(NA,nrow=I*J,ncol=n)
k <- 1
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desb[,j])==2,]
XK12[k,] <- apply(sub,2,mean)
k <- k+1
}
}
XK13 <- matrix(NA,nrow=I*H,ncol=n)
k <- 1
for (h in seq_len(H)) {
for (i in seq_len(I)) {
sub <- X[(Desa[,i]+Desc[,h])==2,]
XK13[k,] <- apply(sub,2,mean)
k <- k+1
}
}
XK23 <- matrix(NA,nrow=J*H,ncol=n)
k <- 1
for (h in seq_len(H)){
for (j in seq_len(J)){
sub <- X[(Desb[,j]+Desc[,h])==2,]
XK23[k,] <- apply(sub,2,mean)
k <- k+1
}
}
NK <- matrix(NA,nrow=I,ncol=J*H)
XK <- matrix(NA,nrow=I*J*H,ncol=n)
k <- 1
for (h in seq_len(H)) {
for (j in seq_len(J)) {
for (i in seq_len(I)) {
sub <- as.matrix(rbind(X[(Desa[,i]+Desb[,j]+Desc[,h])==3,]))
NK[i,(h-1)*J+j] <- sqrt(nrow(sub))
XK[k,] <- apply(sub,2,mean)+XK1[i,]+
XK2[j,]+XK3[h,]-XK12[(j-1)*I+i,]-XK13[(h-1)*I+i,]-
XK23[(h-1)*J+j,]
k <- k+1
}
}
}
XKw <- XK*(as.numeric(NK))
PCA <- PCA.GENES(XKw)
scw <- PCA$scores[,seq_len(Fac)]
ld <- PCA$loadings[,seq_len(Fac)]
ssq <- PCA$var.exp
if (Fac==1) {
scw <- as.matrix(scw)
ld <- as.matrix(ld)
}
if (Fac==0) {
scw <- as.matrix(rep(0,I*J*H))
ld <- as.matrix(rep(0,n))
}
# Re-weigth the scores
sc <- scw/(as.numeric(NK))
XKrec <- sc%*%t(ld)
Xabc <- NULL
TPabc <- NULL
for (i in seq_len(nrow(X))){
position1 <- which(Desa[i,]==1)
position2 <- which(Desb[i,]==1)
position3 <- which(Desc[i,]==1)
Xabc <- rbind(Xabc,XK[I*(position2-1)+I*J*(position3-1)+position1,])
TPabc <- rbind(TPabc,XKrec[I*(position2-1)+
I*J*(position3-1)+position1,])
}
Eabc <- Xabc-TPabc
#Leverage & SPE
leverage <- apply(ld^2,1,sum)
SPE <- apply(Eabc^2,2,sum)
output <- list(XK,sc,ld,ssq,Xabc,TPabc,Eabc,leverage,SPE)
names(output) <- c("data","scores","loadings","var.exp","X","TP","E",
"leverage","SPE")
output
}
################################################################################
################################################################################
# auxiliar.R
busca <- function (x,S) {
which(S[,1] == x[1] & S[,2] == x[2])
}
int.mult <- function(lista,todos=NULL) {
if (is.null(todos)) {
todos <- unlist(lista)
}
comunes <- todos
for (i in seq_len(length(lista))) {
comunes <- intersect(comunes,lista[[i]])
}
comunes
}
n.menor <- function (x,S1,S2) {
length(which(S1 <= x[1] & S2 <= x[2]))
}
noceros <- function (x,num=TRUE,k=0) {
nn <- length(which(x > k))
if (num) {
nn
}
else {
if(nn > 0) {
which(x > k)
}
else {
NULL
}
}
}
sinceros <- function (datos,k) {
datos <- as.matrix(datos)
datos0 <- as.matrix(datos)
if (is.null(k)) {
mini0 <- min(datos[noceros(datos,num=FALSE,k=0)])
kc <- mini0/2
datos0[datos0 == 0] <- kc
}
else {
datos0[datos0 == 0] <- k
}
datos0
}
sim.samples <- function(counts1,counts2=NULL,pnr=1,nss=5,v=0.02) {
seqdep <- c(sum(counts1),sum(counts2))
num.reads1 <- (pnr + c(-v,v))*seqdep[1]
muestras <- vector("list")
muestras$c1 <- NULL
for (s in seq_len(nss)) {
tama <- round(runif(1,num.reads1[1],num.reads1[2]),0)
muestras$c1 <- cbind(muestras$c1,rmultinom(1,size=tama,prob=counts1))
}
if (!is.null(counts2)) {
num.reads2 <- (pnr + c(-v,v))*seqdep[2]
muestras$c2 <- NULL
for (s in seq_len(nss)) {
tama <- round(runif(1,num.reads2[1],num.reads2[2]),0)
muestras$c2 <- cbind(muestras$c2,
rmultinom(1,size=tama,prob=counts2))
}
}
muestras
}
ranking <- function(results) {
M <- results$M
D <- results$D
prob <- results$prob
## Changing NA by 0
M[is.na(M)] <- 0
D[is.na(D)] <- 0
prob[is.na(prob)] <- 0
## Ranking
ranking5 <- sqrt(M*M + D*D)*sign(M)
## Ranking results
theranking <- data.frame(rownames(results),ranking5)
rownames(theranking) <- NULL
colnames(theranking) <- c("ID","statistic")
theranking
}
plot.y2 <- function(x,yright,yleft,yrightlim=range(yright,na.rm=TRUE),
yleftlim=range(yleft,na.rm=TRUE),xlim=range(x,na.rm=TRUE),xlab=NULL,
yylab=c("",""),lwd=c(2,2),pch=c(1,2),col=c(1,2),type=c("o","o"),
linky=TRUE,smooth=0,bg=c("white","white"),lwds=1,length=10,...,x2=NULL,
yright2=NULL,yleft2=NULL,col2=c(3,4)) {
## Plotting RIGHT axis data
plot(x,yright,axes=FALSE,ylab="",xlab=xlab,ylim=yrightlim,xlim=xlim,
pch=pch[1],type=type[1],lwd=lwd[1],col=col[1],...)
axis(4,pretty(yrightlim,length),col=1,col.axis=1)
if (is.null(yright2) == FALSE) {
points(x2,yright2,type=type[1],pch=pch[1],lwd=lwd[1],col=col2[1],...)
}
if (smooth != 0)
lines(supsmu(x,yright,span=smooth),col=col[1],lwd=lwds,...)
if (yylab[1]=="") {
mtext(deparse(substitute(yright)),side=4,outer=FALSE,line=2,col=1,...)
}
else {
mtext(yylab[1],side=4,outer=FALSE,line=2,col=1,...)
}
par(new=TRUE)
## Plotting LEFT axis data
plot(x,yleft,axes=FALSE,ylab="" ,xlab=xlab,ylim=yleftlim,xlim=xlim,bg=bg[1],
pch=pch[2],type=type[2],lwd=lwd[2],col=col[2],...)
box()
axis(2,pretty(yleftlim,length),col=1,col.axis=1)
if (is.null(yleft2) == FALSE) {
points(x2,yleft2,type=type[2],pch=pch[2],bg=bg[2],lwd=lwd[2],
col=col2[2],...)
}
if (smooth != 0)
lines(supsmu(x,yleft,span=smooth),col=col[2],lwd=lwds,...)
if(yylab[2] == "") {
mtext(deparse(substitute(yleft)),side=2,outer=FALSE,line=2,col=1,...)
}
else {
mtext(yylab[2],side=2,outer=FALSE,line=2,col=1,...)
}
## X-axis
axis(1,at=pretty(xlim,length))
}
logscaling <- function(data,base=2,k=1) {
logmaximo <- round(max(data,na.rm=TRUE),0)
numceros <- nchar(logmaximo)-1
etiquetas <- c(0,10^(seq_len(numceros)))
donde <- log(etiquetas + k,base=base)
data <- log(data + k,base=base)
list("data"=data,"at"=donde,"labels"=etiquetas)
}
miscolores <- colors()[c(554,89,111,512,17,586,132,428,601,568,86,390)]
################################################################################
################################################################################
# biodetection.plot.R
biodetection.dat <- function(input,factor=NULL,k=0) {
if (inherits(input,"eSet") == FALSE)
stop("Error. The input data must be an eSet object\n")
if (any(!is.na(featureData(input)@data$Biotype)) == FALSE)
stop ("No biological classification was provided.\nPlease run ",
"addData() function to add this information\n")
if (!is.null(assayData(input)$exprs)) {
dat <- as.matrix(assayData(input)$exprs)
mysamples <- colnames(assayData(input)$exprs)
}
else {
dat <- as.matrix(assayData(input)$counts)
mysamples <- colnames(assayData(input)$counts)
}
numgenes <- nrow(dat)
if (is.null(factor)) { # per sample
cat("Biotypes detection is to be computed for:\n")
print(colnames(dat))
biotablas <- vector("list",length=NCOL(dat))
names(biotablas) <- colnames(dat)
}
else { # per condition
mifactor <- pData(input)[,factor]
niveles <- levels(mifactor)
cat("Biotypes detection is to be computed for:\n")
print(niveles)
biotablas <- vector("list",length=length(niveles))
names(biotablas) <- niveles
dat <- vapply(niveles,
function (k) rowSums(as.matrix(dat[,grep(k,mifactor)])),numeric(1))
}
infobio <- as.character(featureData(input)@data$Biotype)
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
for (i in seq_len(length(biotablas))) {
detect <- dat[,i] > k
perdet1 <- genome*table(infobio,detect)[names(genome),"TRUE"]/
table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(genome),"TRUE"] /
sum(table(infobio,detect)[,"TRUE"])
biotablas[[i]] <- as.matrix(rbind(perdet1[ordre],perdet2[ordre]))
rownames(biotablas[[i]]) <- c("detectionVSgenome","detectionVSsample")
}
mybiotable <- list("genome"=genome[ordre],"biotables"=biotablas,
"genomesize"=numgenes)
mybiotable
}
biodetection.plot <- function(dat,samples=c(1,2),
plottype=c("persample","comparison"),toplot="protein_coding",
toreport=FALSE,...) {
mypar=par(no.readonly=TRUE)
plottype=match.arg(plottype)
if (length(samples) > 2) {
stop("ERROR: This function cannot generate plots for more than 2 ",
"samples.\nPlease,use it as many times as needed to generate the ",
"plots for all your samples.\n")
}
if (is.numeric(samples))
samples=names(dat$biotables)[samples]
biotable1 <- rbind(dat$genome,dat$biotables[[samples[1]]],
rep(0,length(dat$genome)))
# Computing ylim for left and right axis
if (ncol(biotable1) >= 3) {
ymaxL <- ceiling(max(biotable1[,c(1,2,3)],na.rm=TRUE))
ymaxR <- max(biotable1[,-c(1,2,3)],na.rm=TRUE)
}
else {
ymaxL <- ceiling(max(biotable1,na.rm=TRUE))
ymaxR=0
}
if (length(samples) == 2) {
biotable2 <- rbind(dat$genome,dat$biotables[[samples[2]]],
rep(0,length(dat$genome)))
if (ncol(biotable2) >= 3) {
ymax2 <- ceiling(max(biotable2[,c(1,2,3)],na.rm=TRUE))
ymax2sin <- max(biotable2[,-c(1,2,3)],na.rm=TRUE)
ymaxR <- ceiling(max(ymaxR,ymax2sin))
}
else {
ymax2 <- ceiling(max(biotable2,na.rm=TRUE))
}
ymaxL=max(ymaxL,ymax2)
}
# Rescaling biotables (datos2)
if (length(samples) == 2) {
if (ncol(biotable2) >= 3)
biotable2[,-c(1,2,3)] <- biotable2[,-c(1,2,3)]*ymaxL/ymaxR
}
# Rescaling biotables (datos1)
if (ncol(biotable1) >= 3)
biotable1[,-c(1,2,3)] <- biotable1[,-c(1,2,3)]*ymaxL/ymaxR
## PLOTS
if (length(samples) == 1) { # Plot (1 sample) - 2 scales
par(mar=c(11,4,2,2))
barplot(biotable1[c(1,3),],main=samples[1],xlab=NULL,ylab="%features",
axis.lty=1,legend=FALSE,beside=TRUE,col=c("grey",2),las=2,
ylim=c(0,ymaxL),border=c("grey",2))
barplot(biotable1[c(2,4),],main=samples[1],xlab=NULL,ylab="%features",
axis.lty=1,legend=FALSE,beside=TRUE,col=c(2,1),las=2,density=30,
ylim=c(0,ymaxL),border=2,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("% in genome","detected","% in sample"))
}
else { # Plot (2 samples)
par(mar=c(11,4,2,2))
if (plottype == "persample") {
barplot(biotable1[c(1,3),],main=samples[1],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",2),las=2,ylim=c(0,ymaxL),border=c("grey",2))
barplot(biotable1[c(2,4),],main=samples[1],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c(2,1),las=2,density=30,ylim=c(0,ymaxL),border=2,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("% in genome","detected","% in sample"))
# Datos2
barplot(biotable2[c(1,3),],main=samples[2],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",4),las=2,ylim=c(0,ymaxL),border=c("grey",4))
barplot(biotable2[c(2,4),],main=samples[2],xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,col=c(4,1),
las=2,density=30,ylim=c(0,ymaxL),border=4,add=TRUE)
# if number of biotypes >= 3 so we have left and right axis
if (ymaxR > 0) {
axis(side=4,at=pretty(c(0,ymaxL),n=5),
labels=round(pretty(c(0,ymaxL),n=5)*ymaxR/ymaxL,1))
abline(v=9.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("% in genome","detected","% in sample"))
}
# A plot comparing two samples with regard to genome and for % in sample
if (plottype == "comparison") {
lefttable <- rbind(100*dat$biotables[[samples[1]]][1,]/dat$genome,
100*dat$biotables[[samples[2]]][1,]/dat$genome)
righttable <- rbind(dat$biotables[[samples[1]]][2,],
dat$biotables[[samples[2]]][2,])
if (length(toplot) > 1) {
toplot <- toplot[1]
print("WARNING: More than one biotype was provided, the ",
"proportion test will only by applied to the first ",
"biotype.")
}
if ((toplot != 1) && (toplot != "global")) {
numgenes <- dat$genomesize
myx <- round(righttable[,toplot]*numgenes/100,0)
mytest <- prop.test(x=myx,n=rep(numgenes,2),
alternative="two.sided")
if (is.numeric(toplot))
toplot <- colnames(righttable)[toplot]
}
asumar <- colSums(righttable)
asumar <- which(asumar < 0.25)
if (length(asumar) > 1) {
righttable <- cbind(righttable[,-asumar],
rowSums(righttable[,asumar]))
colnames(righttable)[ncol(righttable)] <- "Others"
}
# Detection in the genome
bbb <- barplot(lefttable,main="Biotype detection over genome total",
xlab=NULL,ylab="% detected features",axis.lty=1,legend=FALSE,
cex.names=0.8,beside=TRUE,col=c(2,4),las=2,density=80,
border=c(2,4),ylim=c(0,100))
bbb <- colSums(bbb)/2
lines(bbb,dat$genome,pch=20,type="o",lwd=2)
# %detection in the sample
barplot(righttable,main="Relative biotype abundance in sample",
xlab=NULL,ylab="Relative % biotypes",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(2,4),las=2,border=c(2,4))
legend(x="topright",bty="n",horiz=FALSE,pch=c(15,15,20),
lwd=c(NA,NA,1),legend=c(samples,"% in genome"),col=c(2,4,1))
if ((toplot != 1) && (toplot != "global")) {
print(paste("Percentage of",toplot,"biotype in each sample:"))
names(mytest$estimate) <- samples
print(round(mytest$estimate*100,4))
print(paste("Confidence interval at 95% for the difference of",
"percentages:",samples[1],"-",samples[2]))
print(round(mytest$conf.int[c(1,2)]*100,4))
if (mytest$p.value < 0.05) {
print(paste("The percentage of this biotype is",
"significantly DIFFERENT for these two samples",
"(p-value =",signif(mytest$p.value,4),")."))
}
else {
print(paste("The percentage of this biotype is NOT",
"significantly different for these two samples",
"(p-value =",signif(mytest$p.value,4),")."))
}
}
}
}
# Reset with the default values
if (!toreport)
par(mypar)
}
################################################################################
################################################################################
# cd.plot.R
cd.dat <- function (input,norm=FALSE,refColumn=1) {
if (inherits(input,"eSet") == FALSE)
stop("ERROR: The input data must be an eSet object.\n")
if (!is.null(assayData(input)$exprs)) {
if (ncol( assayData(input)$exprs) < 2)
stop("ERROR: The input object should have at least two samples.\n")
datos <- assayData(input)$exprs
}
else {
if (ncol(assayData(input)$counts) < 2)
stop("ERROR: The input object should have at least two samples.\n")
datos <- assayData(input)$counts
}
ceros = which(rowSums(datos) == 0)
hayceros = (length(ceros) > 0)
if (hayceros) {
print(paste("Warning:",length(ceros),"features with 0 counts in all",
"samples are to be removed for this analysis."))
datos <- datos[-ceros,]
}
## scaling data and/or changing 0 to k
if (norm)
datos <- sinceros(datos, k = NULL)
else
datos <- noirpkm(datos,long=1000,lc=1,k=0.5)
## to plot
data2plot <- log2(datos/datos[,refColumn])
if (is.numeric(refColumn))
refColumn <- colnames(datos)[refColumn]
print(paste("Reference sample is:",refColumn))
MsinRef <- as.matrix(data2plot[,-match(refColumn, colnames(data2plot))])
colnames(MsinRef) <- colnames(data2plot)[-match(refColumn,
colnames(data2plot))]
alpha <- 0.05
alpha <- alpha/ncol(MsinRef)
nperm <- 10^3
bootmed <- vapply(seq_len(nperm),function(k) {
permut <- sample(seq_len(nrow(MsinRef)),replace=TRUE,nrow(MsinRef))
permut <- as.matrix(MsinRef[permut,])
permut <- apply(permut,2,median)
permut
},numeric(ncol(MsinRef)))
if (is.null(dim(bootmed)))
bootmed <- t(as.matrix(bootmed))
bootmed <- t(apply(bootmed,1,quantile,probs=round(c(alpha/2,1-alpha/2),4)))
diagno <- apply(bootmed,1,function (x) {
ddd <- (x[1] <= 0) * (0 <= x[2])
if (ddd == 1)
ddd = "PASSED"
else
ddd = "FAILED"
ddd
})
bootmed <- cbind(bootmed,diagno)
rownames(bootmed) <- colnames(MsinRef)
colnames(bootmed)[3] <- "Diagnostic Test"
print("Confidence intervals for median of M:")
print(bootmed)
if ("FAILED" %in% bootmed[,3])
print(paste0("Diagnostic test: FAILED. Normalization is required ",
"to correct this bias."))
else
print("Diagnostic test: PASSED.")
list("data2plot"=data2plot,"refColumn"=refColumn,"DiagnosticTest"=bootmed)
}
cd.plot <- function (dat,samples=NULL,...) {
refColumn <- dat$refColumn
dat <- dat$data2plot
if (is.null(samples))
samples <- seq_len(ncol(dat))
if (is.numeric(samples))
samples <- colnames(dat)[samples]
samples <- setdiff(samples,refColumn)
if(length(samples) > 12)
stop("Please select 12 samples or less to be plotted (excluding ",
"reference).")
dat <- as.matrix(dat[,samples])
dat.dens <- apply(dat,2,density,adjust=1.5)
limY <- c(0,max(vapply(dat.dens,
function(x) max(x$y,na.rm=TRUE),numeric(1))))
plot(dat.dens[[1]],xlab="M = log2(sample/refsample)",ylab="Density",
lwd=2,ylim=limY,type="l",col=miscolores[1],
main=paste("Reference sample:",refColumn),...)
abline(v=median(dat[,1],na.rm=TRUE),col=miscolores[1],lty=2)
if (length(samples) > 1) {
for (i in 2:length(samples)) {
lines(dat.dens[[i]],col=miscolores[i],lwd=2)
abline(v=median(dat[,i],na.rm=TRUE),col=miscolores[i],lty=i+1)
}
}
legend("topleft",legend=samples,
text.col=miscolores[seq_len(length(samples))],bty="n",lty=1,lwd=2,
col=miscolores[seq_len(length(samples))])
}
################################################################################
################################################################################
# countsbio.plot.R
countsbio.dat <- function (input,biotypes=NULL,factor=NULL,norm=FALSE) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an eSet object\n")
if (!is.null(assayData(input)$exprs))
datos <- assayData(input)$exprs
else
datos <- assayData(input)$counts
depth <- round(colSums(datos)/10^6,1)
names(depth) <- colnames(datos)
ceros <- which(rowSums(datos) == 0)
hayceros <- length(ceros) > 0
if (hayceros) {
warning("Warning:",length(ceros),"features with 0 counts in all ",
"samples are to be removed for this analysis.")
datos0=datos[-ceros,,drop=FALSE]
}
else {
datos0=datos
}
nsam <- NCOL(datos)
if (nsam == 1) {
datos <- as.matrix(datos)
datos0 <- as.matrix(datos0)
}
# Per condition
if (is.null(factor)) { # per sample
print("Count distributions are to be computed for:")
print(colnames(datos))
}
else { # per condition
mifactor <- pData(input)[,factor]
niveles <- levels(mifactor)
print("Counts per million distributions are to be computed for:")
print(niveles)
if (norm) {
datos <- vapply(niveles,function (k) {
rowMeans(as.matrix(datos[,grep(k,mifactor)]))
},numeric(1))
datos0 <- vapply(niveles,function (k) {
rowMeans(as.matrix(datos0[,grep(k,mifactor)]))
},numeric(1))
}
else {
datos <- vapply(niveles,function (k) {
10^6*rowMeans(t(t(datos[,grep(k,mifactor)])/colSums(as.matrix(
datos[,grep(k,mifactor)]))))
},numeric(1))
datos0 <- vapply(niveles,function (k) {
10^6*rowMeans(t(t(datos0[,grep(k,mifactor)])/colSums(as.matrix(
datos0[,grep(k,mifactor)]))))
},numeric(1))
}
colnames(datos) <- colnames(datos0) <- niveles
depth <- vapply(niveles,function (k) {
paste(range(depth[grep(k,mifactor)]),collapse="-")
},character(1))
}
# Biotypes
if (!is.null(featureData(input)$Biotype)) {
if (hayceros) {
infobio0 <- as.character(featureData(input)$Biotype)[-ceros]
}
else {
infobio0 <- as.character(featureData(input)$Biotype)
}
infobio <- as.character(featureData(input)$Biotype)
}
else {
infobio0 <- NULL
infobio <- NULL
}
if (!is.null(infobio)) {
if(is.null(biotypes)) {
biotypes <- unique(infobio)
names(biotypes) <- biotypes
}
# which genes belong to each biotype
biog <- lapply(biotypes,function(x) {
which(is.element(infobio0,x))
})
names(biog) <- biotypes
bionum <- c(NROW(datos0),vapply(biog,length,integer(1)))
names(bionum) <- c("global",names(biotypes))
bio0 <- which(bionum == 0)
if (length(bio0) > 0)
bionum <- bionum[-bio0]
}
else {
biotypes <- NULL
bionum <- NULL
}
# Create the summary matrix information
if (is.null(bionum)) {
resumen <- vector("list",length=1)
names(resumen) <- "global"
}
else {
resumen <- vector("list",length=length(bionum))
names(resumen) <- names(bionum)
}
cuentas <- c(0,1,2,5,10)
if (is.null(factor)) {
if (norm) {
datosCPM <- datos
}
else {
datosCPM=10^6 * t(t(datos)/colSums(as.matrix(datos)))
}
} else { datosCPM=datos }
for (i in seq_len(length(resumen))) {
if (i == 1) {
datosR=datosCPM
} else {
if(!is.null(infobio)) {
datosR <- datosCPM[which(infobio == names(resumen)[i]),]
if (!is(datosR,"matrix")) {
datosR <- t(as.matrix(datosR))
}
}
}
nfeatures <- nrow(datosR)
datosR <- datosR[which(rowSums(datosR) > 0),]
if (!is(datosR,"matrix")) {
datosR <- t(as.matrix(datosR))
}
myglobal <- NULL
mypersample <- NULL
for (kk in seq_len(length(cuentas))) {
mypersample <- rbind(mypersample,apply(datosR,2,function (x) {
length(which(x > cuentas[kk]))
}))
myglobal <- c(myglobal,sum(apply(datosR,1,function (x) {
max(x) > cuentas[kk]
})))
}
mypersample <- round(100*mypersample/nfeatures,1)
mypersample <- rbind(mypersample,depth)
rownames(mypersample) <- seq_len(nrow(mypersample))
myglobal <- c(round(100*myglobal/nfeatures,1),nfeatures)
resumen[[i]] <- data.frame(c(paste("CPM >",cuentas),"depth"),
mypersample,"total"=myglobal)
colnames(resumen[[i]])[1] <- names(resumen)[i]
colnames(resumen[[i]])[2:(ncol(resumen[[i]])-1)] <- colnames(datosR)
}
## results
cosas <- list("result"=datos0,"bionum"=bionum,"biotypes"=infobio0,
"summary"=resumen)
cosas
}
countsbio.plot <- function(dat,samples=c(1,2),toplot="global",
plottype=c("barplot","boxplot"),toreport=FALSE,...) {
# dat: Data coming from countsbio.dat function
# samples: Samples to be plotted. If NULL,all samples are plotted.
# toplot: Name of biotype (including "global") to be plotted.
mypar <- par(no.readonly=TRUE)
plottype <- match.arg(plottype)
## Preparing data
if (is.null(samples)) {
if (NCOL(dat$result) == 1) {
samples=1
} else {
samples <- seq_len(NCOL(dat$result))
}
}
if (is.numeric(toplot))
toplot <- names(dat$summary)[toplot]
if (is.numeric(samples) && !is.null(colnames(dat$result)))
samples <- colnames(dat$result)[samples]
if (plottype == "barplot") {
if ((exists("ylab") && !is.character(ylab)) || !exists("ylab"))
ylab <- ""
datos <- dat$summary[[toplot]]
mytotal <- as.numeric(datos[,"total"])
datos <- as.matrix(datos[,samples])
rownames(datos) <- as.character(dat$summary[[toplot]][,1])
par(mar=c(6,4,4,2))
barplot(as.numeric(datos[1,]),col=miscolores[1],las=2,main="",ylab="",
density=70,ylim=c(0,100),cex.axis=0.8,names.arg="",...)
for (i in 2:(length(mytotal)-2)) {
barplot(as.numeric(datos[i,]),col=miscolores[i],las=2,main="",
ylab="",add=TRUE,density=70,ylim=c(0,100),cex.axis=0.8,
names.arg="",...)
}
bp <- barplot(as.numeric(datos[(length(mytotal)-1),]),
col=miscolores[(length(mytotal)-1)],las=2,
main=paste(toupper(toplot)," (",mytotal[length(mytotal)],")",
sep=""),ylab="Sensitivity (%)",add=TRUE,names.arg=colnames(datos),
cex.axis=0.8,density=70,ylim=c(0,100),cex.names=0.8,...)
for (j in seq_len((length(mytotal)-1)))
abline(h=mytotal[j],col=miscolores[j],lwd=2)
if (length(samples) <= 10) {
mtext(side=3,text=datos["depth",],adj=0.5,at=bp,cex=0.8)
}
else {
mtext(side=3,text=datos["depth",],at=bp,cex=0.7,las=2)
}
legend("top",rownames(datos)[-length(mytotal)],fill=miscolores,
density=70,bty="n",ncol=3)
par(mar=c(5,4,4,4) + 0.1)
}
if (plottype == "boxplot") {
conteos <- as.matrix(dat$result[,samples])
if (is.numeric(samples))
colnames(conteos) <- colnames(dat$result)[samples]
else
colnames(conteos) <- samples
num <- dat$bionum[toplot]
if (is.null(num)) {
if (toplot == "global") {
num=nrow(conteos)
} else {
num=0
}
}
infobio <- dat$biotypes
if (num == 0 && toplot != "global")
stop("Error: No data available. Please,change toplot parameter.")
if ((exists("ylab") && !is.character(ylab)) || !exists("ylab"))
ylab <- "Expression values"
## Plots
if (length(samples) == 1) {
escala <- logscaling(conteos,base=2)
if (is.null(infobio)) {
boxplot(escala$data,col=miscolores[1],ylab=ylab,main="",
yaxt="n",...)
}
else {
par(mar=c(10,4,4,2))
boxplot(escala$data ~ infobio,col=miscolores,ylab=ylab,
main=colnames(conteos),las=2,cex.axis=0.8,cex.lab=0.9,
yaxt="n",...)
cuantos <- dat$bionum[-1]
cuantos <- cuantos[sort(names(cuantos))]
mtext(cuantos,3,at=seq_len(length(cuantos)),cex=0.6,las=2)
}
}
else {
if (toplot != "global")
conteos=conteos[which(infobio == toplot),]
escala <- logscaling(conteos,base=2)
main <- paste(toupper(toplot)," (",num,")",sep="")
par(mar=c(6,4,2,2))
boxplot(escala$data,col=miscolores,ylab=ylab,main=main,las=2,
cex.lab=0.9,cex.axis=0.8,yaxt="n",...)
}
axis(side=2,at=escala$at,labels=escala$labels)
}
if (!toreport)
par(mypar)
}
################################################################################
################################################################################
# dat.R
dat <- function(input,type = c("biodetection","cd","countsbio","GCbias",
"lengthbias","saturation","PCA"),k=0,ndepth=6,factor=NULL,norm=FALSE,
refColumn=1,logtransf = FALSE) {
type <- match.arg(type)
if (type == "biodetection") {
output <- new("Biodetection",
dat=biodetection.dat(input,factor=factor,k=k))
}
if (type == "cd") {
output <- new("CD",dat=cd.dat(input,norm=norm,refColumn=refColumn))
}
if (type == "countsbio") {
output <- new("CountsBio",
dat=countsbio.dat(input,factor=factor,norm=norm))
}
#if (type == "GCbias") {
# output <- new("GCbias",
# dat=GC.dat(input,factor=factor,norm=norm))
#}
#if (type == "lengthbias") {
# output <- new("lengthbias",
# dat=length.dat(input,factor=factor,norm=norm))
#}
if (type == "saturation") {
output <- new("Saturation",dat=saturation.dat(input,k=k,ndepth=ndepth))
}
#if (type == "PCA") {
# output=new("PCA",dat=PCA.dat(input,norm=norm,logtransf=logtransf))
#}
output
}
################################################################################
################################################################################
# DE.plot.R
DE.plot <- function(output,q=NULL,graphic=c("MD","expr","chrom","distr"),
pch=20,cex=0.5,col=1,pch.sel=1,cex.sel=0.6,col.sel=2,log.scale=TRUE,
chromosomes=NULL,join=FALSE,...) {
mypar <- par(no.readonly=TRUE)
if (!is(output,"Output"))
stop("Error. Output argument must contain an object generated by ",
"noiseq or noiseqbio functions.\n")
graphic <- match.arg(graphic)
noiseqbio <- "theta" %in% colnames(output@results[[1]])[c(1,2,3,4)]
## MD plot
if (graphic == "MD") {
if (noiseqbio) {
M <- output@results[[1]][,"log2FC"]
D <- abs(output@results[[1]][,1] - output@results[[1]][,2])+1
names(M) <- names(D) <- rownames(output@results[[1]])
plot(M,D,pch=pch,xlab="M",ylab="D",cex=cex,col=col,log="y",...)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(M[mySelection],D[mySelection],col=col.sel,pch=pch.sel,
cex=cex.sel)
}
}
else {
plot(output@results[[1]][,"M"],(1+output@results[[1]][,"D"]),
pch=pch,xlab="M",ylab="D",cex=cex,col=col,log="y",...)
if (!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(output@results[[1]][mySelection,"M"],
output@results[[1]][mySelection,"D"]+1,col=col.sel,
pch=pch.sel,cex=cex.sel)
}
}
}
## Expression plot: Condition1 vs Condition2
else if (graphic == "expr") {
data <- cbind(output@results[[1]][1],output@results[[1]][2])
rownames(data) <- rownames(output@results[[1]])
colnames(data) <- colnames(output@results[[1]][c(1,2)])
if (log.scale) {
k <- min(data,na.rm=TRUE)
escala <- logscaling(data,base=2,k=k)
plot(escala$data,pch=pch,cex=cex,col=col,yaxt="n",xaxt="n",...)
axis(side=1,at=escala$at,labels=escala$labels)
axis(side=2,at=escala$at,labels=escala$labels)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(escala$data[mySelection,],col=col.sel,pch=pch.sel,
cex=cex.sel)
}
}
else {
plot(data,pch=pch,cex=cex,col=col,...)
if(!is.null(q)) {
mySelection <- rownames(degenes(output,q))
points(data[mySelection,],col=col.sel,pch=pch.sel,cex=cex.sel)
}
}
}
## MANHATTAN PLOT
else if (graphic == "chrom") {
mydata <- data.frame(as.character(output@results[[1]][,"Chrom"]),
output@results[[1]][,c("GeneStart","GeneEnd")],
output@results[[1]][,c(1,2)])
mydata <- na.omit(mydata)
colnames(mydata) <- c("chr","start","end",colnames(mydata)[-c(1,2,3)])
if (is.null(chromosomes)) { # todos los cromosomas
chromosomes <- unique(mydata$chr)
chromosomes <- sort(chromosomes)
}
print("REMEMBER. You are plotting these chromosomes and in this order:")
print(chromosomes)
# logarithmic scale
if (log.scale) {
if (min(mydata[,-c(1,2,3)],na.rm=TRUE) < 1) {
kk <- -min(mydata[,-c(1,2,3)],na.rm=TRUE)+1
}
else {
kk <- 0
}
mydata[,-c(1,2,3)] <- log2(mydata[,-c(1,2,3)]+kk)
}
# Selecting chromosomes and ordering positions
ordenat <- NULL
for (cromo in chromosomes) {
myselec <- mydata[mydata[,"chr"] == cromo,]
myselec <- myselec[order(myselec[,"start"]),]
ordenat <- rbind(ordenat,myselec)
}
sel.ord <- NULL
if (!is.null(q)) {
# up-regulated in first condition
cond1 <- rownames(degenes(output,q,M="up"))
# up-regulated in second condition
cond2 <- rownames(degenes(output,q,M="down"))
cond1 <- (rownames(ordenat) %in% cond1)*1
cond2 <- (rownames(ordenat) %in% cond2)*1
sel.ord <- cbind(cond1,cond2)
rownames(sel.ord) <- rownames(ordenat)
}
chr.long <- aggregate(ordenat[,"end"],
by=list(as.character(ordenat[,"chr"])),max)
chr.long <- chr.long[match(chromosomes,chr.long[,1]),]
if (join) { # si todos los cromosomas van en el mismo plot
total.long <- sum(as.numeric(chr.long$x))
chr.start <- cumsum(c(1,chr.long$x[-length(chr.long$x)]))
names(chr.start) <- chr.long[,1]
plot(c(1,total.long),c(-max(ordenat[,5]),max(ordenat[,4])),
type="n",xlab="",ylab="Expression data",xaxt="n")
axis(side=1,at=chr.start,labels=chr.long[,1],font=2)
abline(h=0,lty=2,lwd=0.5)
for (ch in chromosomes) {
dat.chr <- ordenat[which(ordenat[,"chr"] == ch),]
dat.chr[,c("start","end")] <- dat.chr[,c("start","end")] +
chr.start[ch] - 1
rect(xleft=dat.chr[,"start"],ybottom=0,xright=dat.chr[,"end"],
ytop=dat.chr[,4],col="grey",border=NA)
rect(xleft=dat.chr[,"start"],ybottom=-dat.chr[,5],
xright=dat.chr[,"end"],ytop=0,col="grey",border=NA)
if (!is.null(q)) {
aux <- which(rownames(sel.ord) %in% rownames(dat.chr))
sel.chr1 <- dat.chr[,4]*sel.ord[aux,1]
sel.chr2 <- -dat.chr[,5]*sel.ord[aux,2]
rect(xleft=dat.chr[,"start"],ybottom=0,
xright=dat.chr[,"end"],ytop=sel.chr1,col=2,border=NA)
rect(xleft=dat.chr[,"start"],ybottom=sel.chr2,
xright=dat.chr[,"end"],ytop=0,col=3,border=NA)
}
segments(x0=dat.chr[,"start"],y0=0,x1=dat.chr[,"end"],y1=0,
col=4,lwd=0.5) # annotated genes
}
text(c(1,1),0.9*c(max(ordenat[,4]),-max(ordenat[,5])),
colnames(mydata)[4:5],font=3,adj=0,col=2:3)
# a plot for each chromosome
}
else {
num.chr <- length(chromosomes)
k <- 20
long.prop <- round((chr.long$x / min(chr.long$x))*k,0)
while (max(long.prop)*num.chr > 500 | max(long.prop) > 50) {
k <- k-1
long.prop <- round((chr.long$x / min(chr.long$x))*k,0)
}
forlayout <- matrix(0,num.chr,max(long.prop))
for (i in seq_len(num.chr)) {
forlayout[i,seq_len(long.prop[i])] <- i
}
layout(forlayout)
if (num.chr > 1)
par(mar=c(2,4.1,0.1,0.1))
miylim <- c(-max(ordenat[,5],na.rm=TRUE),
max(ordenat[,4],na.rm=TRUE))
for (i in seq_len(num.chr)) {
apintar <- ordenat[which(ordenat[,"chr"] == chromosomes[i]),2:5]
plot(c(1,chr.long[i,2]),miylim,type="n",xlab="",
ylab=chromosomes[i],xaxt="n",font.lab=2,cex.lab=1.3)
rect(xleft=apintar[,"start"],ybottom=0,xright=apintar[,"end"],
ytop=apintar[,3],col="grey",border=NA)
rect(xleft=apintar[,"start"],ybottom=-apintar[,4],
xright=apintar[,"end"],ytop=0,col="grey",border=NA)
if (!is.null(q)) {
aux <- which(rownames(sel.ord) %in% rownames(apintar))
asel <- sel.ord[aux,]*apintar[,3:4]
rect(xleft=apintar[,"start"],ybottom=0,
xright=apintar[,"end"],ytop=asel[,1],col=2,border=NA)
rect(xleft=apintar[,"start"],ybottom=-asel[,2],
xright=apintar[,"end"],ytop=0,col=3,border=NA)
}
abline(h=0,lty=2,lwd=0.5)
segments(x0=apintar[,"start"],y0=0,x1=apintar[,"end"],y1=0,
col=4,lwd=0.5) # annotated genes
etiq <- mypretty(c(1,chr.long[i,2]),n=10)
axis(side=1,at=etiq,labels=etiq)
text(c(1,1),0.9*miylim,colnames(mydata)[5:4],font=3,adj=0,
col=3:2)
}
}
}
## DEG distribution across biotypes/chromosomes
else if (graphic == "distr") {
if(!is.null(q)) { # Computing DEG
mySelection <- rownames(degenes(output,q))
detect <- rownames(output@results[[1]]) %in% mySelection
}
else {
stop("You must specify a valid value for q\n")
}
if ("Chrom" %in% colnames(output@results[[1]])) {
numplot <- 1
infobio <- output@results[[1]][,"Chrom"]
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
perdet1 <- genome*table(infobio,detect)[names(
genome),2]/table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(
genome),2]/sum(table(infobio,detect)[,2])
ceros <- rep(0,length(genome))
biotable1 <- as.matrix(rbind(genome[ordre],perdet1[ordre],
perdet2[ordre],ceros))
rownames(biotable1) <- c("genome","degVSgenome","deg","ceros")
if (!is.null(chromosomes))
biotable1=biotable1[,chromosomes]
ymaxL1 <- ceiling(max(biotable1,na.rm=TRUE))
}
else {
numplot <- 0
}
if ("Biotype" %in% colnames(output@results[[1]])) {
numplot <- c(numplot,1)
infobio <- output@results[[1]][,"Biotype"]
genome <- 100*table(infobio)/sum(table(infobio))
ordre <- order(genome,decreasing=TRUE)
perdet1 <- genome*table(infobio,detect)[names(
genome),2]/table(infobio)[names(genome)]
perdet2 <- 100*table(infobio,detect)[names(
genome),2]/sum(table(infobio,detect)[,2])
ceros <- rep(0,length(genome))
biotable2 <- as.matrix(rbind(genome[ordre],perdet1[ordre],
perdet2[ordre],ceros))
rownames(biotable2) <- c("genome","degVSgenome","deg","ceros")
higher2 <- which(biotable2[1,] > 2)
lower2 <- which(biotable2[1,] <= 2)
if (length(higher2) > 0) {
ymaxL2 <- ceiling(max(biotable2[,higher2],na.rm=TRUE))
if (length(lower2) > 0) {
ymaxR2 <- ceiling(max(biotable2[,lower2],na.rm=TRUE))
biotable2[,lower2] <- biotable2[,lower2]*ymaxL2/ymaxR2
}
else {
ymaxR2 <- ymaxL2
}
}
else {
ymaxR2 <- ceiling(max(biotable2[,lower2],na.rm=TRUE))
ymaxL2 <- ymaxR2
}
}
else {
numplot <- c(numplot,0)
}
# Plot
if (sum(numplot) == 0)
stop("Biotype or chromosome information is needed for this plot\n")
if (sum(numplot) == 1) { # 1 Plot
par(mar=c(10,4,2,2))
if (numplot[1] == 1) {
barplot(biotable1[c(1,3),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c("grey",2),las=2,ylim=c(0,ymaxL1),
border=c("grey",2))
barplot(biotable1[c(2,4),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(2,1),las=2,density=30,ylim=c(0,ymaxL1),
xborder=2,add=TRUE)
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("%Chrom in genome","%DEG in Chrom",
"%Chrom in DEG"))
}
if (numplot[2] == 1) {
barplot(biotable2[c(1,3),],
main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c("grey",4),las=2,ylim=c(0,ymaxL2),
border=c("grey",4))
barplot(biotable2[c(2,4),],
main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,
beside=TRUE,col=c(4,1),las=2,density=30,ylim=c(0,ymaxL2),
border=4,add=TRUE)
axis(side=4,at=pretty(c(0,ymaxL2),n=5),
labels=round(pretty(c(0,ymaxL2),n=5)*ymaxR2/ymaxL2,1))
if (ymaxR2 != ymaxL2) {
abline(v=3*length(higher2) + 0.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("%Biotype in genome","%DEG in Biotype",
"%Biotype in DEG"))
}
}
if (sum(numplot) == 2) { # 2 Plots
par(mar=c(10,4,2,2),mfrow=c(1,2))
# Chromosomes
barplot(biotable1[c(1,3),],
main="DEG distribution across chromosomes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",2),las=2,ylim=c(0,ymaxL1),border=c("grey",2))
barplot(biotable1[c(2,4),],
main="DEG distribution across chromosomes",xlab=NULL,
ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,col=c(2,1),
las=2,density=30,ylim=c(0,ymaxL1),border=2,add=TRUE)
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",2,2),
density=c(NA,30,NA),border=c("grey",2,2),
legend=c("%Chrom in genome","%DEG in Chrom","%Chrom in DEG"))
# Biotypes
barplot(biotable2[c(1,3),],main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c("grey",4),las=2,ylim=c(0,ymaxL2),border=c("grey",4))
barplot(biotable2[c(2,4),],main="DEG distribution across biotypes",
xlab=NULL,ylab="%features",axis.lty=1,legend=FALSE,beside=TRUE,
col=c(4,1),las=2,density=30,ylim=c(0,ymaxL2),border=4,add=TRUE)
axis(side=4,at=pretty(c(0,ymaxL2),n=5),
labels=round(pretty(c(0,ymaxL2),n=5)*ymaxR2/ymaxL2,1))
if (ymaxR2 != ymaxL2) {
abline(v=3*length(higher2) + 0.5,col=3,lwd=2,lty=2)
}
legend(x="topright",bty="n",horiz=FALSE,fill=c("grey",4,4),
density=c(NA,30,NA),border=c("grey",4,4),
legend=c("%Biotype in genome","%DEG in Biotype",
"%Biotype in DEG"))
}
}
par(mypar)
}
## Auxiliar function
mypretty <- function(x,n=5) {
mywidth <- diff(x)
mybin <- ceiling(mywidth/(n-1))
mylabels0 <- x[1] + mybin*(0:(n-1))
ndig <- nchar(as.character(mylabels0))
mylabels <- mylabels0
k <- 0
for (i in 2:(n-1)) {
mylabels[i] <- (round(mylabels0[i]/10^(ndig[i]-1),k))*10^(ndig[i]-1)
while (mylabels[i] == mylabels[i-1]) {
k <- k+1
mylabels[i] <- (round(mylabels0[i]/10^(ndig[i]-1),k))*10^(ndig[i]-1)
}
}
mylabels
}
################################################################################
################################################################################
# degenes.R
degenes <- function (object,q=0.95,M=NULL) {
if (!is(object,"Output"))
stop("You must give the object returned by the noiseq function\n")
x <- object@results[[1]]
noiseqbio="theta" %in% colnames(x)[seq_len(4)]
if (noiseqbio) {
y <- na.omit(x[c("theta","prob")])
colnames(y)[1]="M"
} else {
y <- na.omit(x[c("M","D","prob")])
}
if (is.null(M)) {
losdeg <- y[y[,"prob"] > q,]
print(paste(dim(losdeg)[1],"differentially expressed features"))
}
else if (M == "up") {
estos <- y[y[,"M"] > 0,]
losdeg <- estos[estos[,"prob"] > q,]
print(paste(dim(losdeg)[1],
"differentially expressed features (up in first condition)"))
}
else if (M == "down") {
estos <- y[y[,"M"] < 0,]
losdeg <- estos[estos[,"prob"] > q,]
print(paste(dim(losdeg)[1],
"differentially expressed features (down in first condition)"))
}
else {
stop("ERROR! Value for parameter M is not valid. Please,choose ",
"between NULL,'up' or 'down'")
}
# Restore the object with the same "results" structure
losdeg <- x[rownames(losdeg),]
losdeg[order(losdeg[,"prob"],decreasing=TRUE),]
}
################################################################################
################################################################################
# fewrepliactes.R
share.info = function (mydata, n1, n2, r, nclust) {
# clustering data by k-means algorithm
# 1.a) Normalized data
gc()
cl <- suppressWarnings(kmeans(mydata,nclust,nstart=25,iter.max=nclust + 30))
cat("...k-means clustering done\n")
cat(paste("Size of",nclust,"clusters:\n"))
print(cl$size)
# Creating pseudo-data
cluster.data <- lapply(seq_len(nclust),function (k) {
mydata[cl$cluster == k,]
})
# Resampling
npermu <- cl$size*r
npermu <- vapply(npermu,function (x) min(x,1000),numeric(1))
cat("Resampling cluster...")
myres=vector("list",length=nclust)
for (i in seq_len(nclust)) {
print(i)
#if (cl$size[i] > 1) { # OPTION 2.A
# OPTION 2.C: small clusters
if (cl$size[i] > 1 && cl$size[i] < 1000) {
#myres[[i]]=t(sapply(seq_len(npermu[i]),function (j) {
# permu=sample(cluster.data[[i]])
# nn1=n1*cl$size[i]
# nn2=n2*cl$size[i]
# mean1=mean(permu[seq_len(nn1)])
# mean2=mean(permu[(nn1+1):(nn1+nn2)])
# sd1=sd(as.numeric(permu[seq_len(nn1)]))
# sd2=sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
# data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
# "M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
# sd2^2/(mean2^2*log(2)^2*nn2)),
# "D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
#}))
myres[[i]]=t(vapply(seq_len(npermu[i]),function (j) {
permu=sample(cluster.data[[i]])
nn1=n1*cl$size[i]
nn2=n2*cl$size[i]
mean1=mean(permu[seq_len(nn1)])
mean2=mean(permu[(nn1+1):(nn1+nn2)])
sd1=sd(as.numeric(permu[seq_len(nn1)]))
sd2=sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
"M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
sd2^2/(mean2^2*log(2)^2*nn2)),
"D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
},vector("list",4)))
}
if (cl$size[i] >= 1000) { # OPTION 2.C & 2.D: big clusters
# Option 2.D: clustering big clusters
cl2 <- kmeans(cluster.data[[i]],nclust,nstart=25,
iter.max=nclust + 20)
cat(paste("Size of",nclust,"subclusters of cluster:",i)); cat("\n")
print(cl2$size)
subcluster.data <- lapply(seq_len(nclust),function (k) {
cluster.data[[i]][cl2$cluster == k,]
})
npermu2 <- cl2$size*r
npermu2 <- vapply(npermu2,function (x) min(x,1000),numeric(1))
## modified to reduce the number of permutations to be done
myres2 <- vector("list",length=nclust)
for (h in seq_len(nclust)) {
if (cl2$size[h] > 1) {
#myres2[[h]] <- t(sapply(seq_len(npermu2[h]),function (j) {
# permu=sample(subcluster.data[[h]])
# nn1 <- n1*cl2$size[h]
# nn2 <- n2*cl2$size[h]
# mean1 <- mean(permu[seq_len(nn1)])
# mean2 <- mean(permu[(nn1+1):(nn1+nn2)])
# sd1 <- sd(as.numeric(permu[seq_len(nn1)]))
# sd2 <- sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
# data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
# "M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
# sd2^2/(mean2^2*log(2)^2*nn2)),
# "D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
#}))
myres2[[h]] <- t(vapply(seq_len(npermu2[h]),function (j) {
permu=sample(subcluster.data[[h]])
nn1 <- n1*cl2$size[h]
nn2 <- n2*cl2$size[h]
mean1 <- mean(permu[seq_len(nn1)])
mean2 <- mean(permu[(nn1+1):(nn1+nn2)])
sd1 <- sd(as.numeric(permu[seq_len(nn1)]))
sd2 <- sd(as.numeric(permu[(nn1+1):(nn1+nn2)]))
data.frame("M"=log2(mean1/mean2),"D"=mean1-mean2,
"M.sd"=sqrt(sd1^2/(mean1^2*log(2)^2*nn1) +
sd2^2/(mean2^2*log(2)^2*nn2)),
"D.sd"=sqrt(sd1^2/sqrt(nn1) + sd2^2/sqrt(nn2)))
},vector("list",4)))
}
}
myres[[i]] <- do.call("rbind",myres2)
}
}
# Computing Zr for noise
cat("Computing Z for noise...\n")
# 4.A) a0: Global for all R*G permutations
myres <- do.call("rbind",myres)
a0.M <- quantile(as.numeric(myres[,"M.sd"]),probs=0.9,na.rm=TRUE)
a0.D <- quantile(as.numeric(myres[,"D.sd"]),probs=0.9,na.rm=TRUE)
M <- as.numeric(myres[,"M"])/(a0.M + as.numeric(myres[,"M.sd"]))
D <- as.numeric(myres[,"D"])/(a0.D + as.numeric(myres[,"D.sd"]))
(M + D)/2
}
################################################################################
# filter.low.counts.R
CV = function(data) {
100*sd(data,na.rm=TRUE)/mean(data,na.rm=TRUE)
}
filtered.data <- function(dataset,factor,norm=TRUE,depth=NULL,method=1,
cv.cutoff=100,cpm=1,p.adj="fdr") {
dataset0=dataset[rowSums(dataset) > 0,]
dataset=dataset0
if ((method == 3) && (norm)) {
if (is.null(depth)) {
stop("ERROR: Sequencing depth for each column in dataset must ",
"be provided.\n")
}
dataset <- t(t(dataset0)/(colSums(dataset0)/depth))
}
if ((method < 3) && (!norm)) {
dataset <- 10^6*t(t(dataset0)/colSums(dataset0))
}
grupos <- unique(factor)
cumple <- NULL
cat("Filtering out low count features...\n")
for (gg in grupos) {
datos <- as.matrix(dataset[,grep(gg,factor)])
if (method == 1) {
if (ncol(datos) == 1) {
cumplecond=(datos > cpm)
}
else {
cumplecond <-
(apply(datos,1,CV) < cv.cutoff)*(rowMeans(datos) > cpm)
cumplecond[which(is.na(cumplecond) == TRUE)] <- 0
}
cumple <- cbind(cumple,cumplecond)
}
if (method == 2) {
if (ncol(datos) == 1)
stop("ERROR: At least 2 replicates per condition are ",
"required to apply this method.")
mytest <- apply(datos,1,function (x) {
suppressWarnings(wilcox.test(x,alternative="greater",
conf.int=FALSE,mu=0))$"p.value"
})
mytest <- p.adjust(mytest,method=p.adj)
cumple <- cbind(cumple,1*(mytest < 0.05))
}
if (method == 3) {
p0=cpm/10^6
mytest <- apply(datos,1,function (x)
suppressWarnings(prop.test(sum(x),n=sum(datos),p=p0,
alternative="greater"))$"p.value")
mytest <- p.adjust(mytest,method=p.adj)
cumple <- cbind(cumple,1*(mytest < 0.05))
}
}
cumple <- which(rowSums(as.matrix(cumple)) >= 1)
cat(paste(length(cumple),"features are to be kept for differential ",
"expression analysis with filtering method",method))
cat("\n")
dataset0[cumple,]
}
################################################################################
################################################################################
# makeASCAdesign.R
make.ASCA.design <- function(x) {
x <- as.factor(x)
levels <- unique(x)
n <- length(x)
p <- length(levels)
Design <- matrix(0,nrow=n,ncol=p)
for (i in seq_len(n)){
for (j in seq_len(p)){
if (x[i]==levels[j]) {
Design[i,j]=1
}
}
}
colnames(Design)<-levels
output<-Design
output
}
################################################################################
################################################################################
# MD.R
MD <- function (dat=dat,selec=c(seq_len(nrow(dat)))) {
pares <- as.matrix(combn(ncol(dat), 2))
if (NCOL(pares) > 30) {
sub30 <- sample(seq_len(NCOL(pares)),size=30,replace=FALSE)
pares <- pares[,sub30]
}
mm <- NULL
dd <- NULL
for (i in seq_len(ncol(pares))) {
a <- dat[selec,pares[1,i]]
b <- dat[selec,pares[2,i]]
mm <- cbind(mm,log(a/b,2))
dd <- cbind(dd,abs(a-b))
}
list("M"=mm,"D"=dd)
}
MDbio <- function (dat=dat,selec= c(seq_len(nrow(dat))),param=NULL,a0per=0.9) {
pares <- as.matrix(combn(ncol(dat), 2))
mm <- NULL
dd <- NULL
for (i in seq_len(ncol(pares))) {
a <- dat[selec,pares[1,i]]
b <- dat[selec,pares[2,i]]
mm <- cbind(mm, log(a/b, 2))
dd <- cbind(dd, (a-b))
}
## Correcting (M,D)
sd.M <- sqrt(param$sd[,1]^2/(dat[,1]^2*log(2)^2*param$n[1]) +
param$sd[,2]^2/(dat[,2]^2*log(2)^2*param$n[2]))
sd.D = sqrt(param$sd[,1]^2/param$n[1] + param$sd[,2]^2/param$n[2])
if(is.null(a0per)) {
a0.M = a0.D = 0
}
else {
if (a0per == "B") {
B <- 100
a0.M <- B*max(sd.M,na.rm=TRUE)
a0.D <- B*max(sd.D,na.rm=TRUE)
}
else {
a0per <- as.numeric(a0per)
a0.M <- quantile(sd.M, probs = a0per, na.rm = TRUE)
a0.D <- quantile(sd.D, probs = a0per, na.rm = TRUE)
}
}
mm <- mm / (a0.M + sd.M)
dd <- dd / (a0.D + sd.D)
# Results
list("M"=mm,"D"=dd)
}
################################################################################
################################################################################
# noiseq.R
noiseq <- function (input,k=0.5,norm=c("rpkm","uqua","tmm","n"),
replicates=c("technical","biological","no"),factor=NULL,conditions=NULL,
pnr=0.2,nss=5,v=0.02,lc=0) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an object generated by the readData ",
"function\n")
if (is.null(factor))
stop("Error. You must specify the factor to know which conditions ",
"you wish to compare. Please,give the argument 'factor'.\n")
replicates <- match.arg(replicates)
if (replicates == "biological")
print("WARNING: Your experiment has biological replicates. You ",
"should consider using NOISeqBIO instead of NOISeq.")
norm <- match.arg(norm)
# (M,D) for signal and noise
print("Computing (M,D) values...")
miMD <- allMD(input,factor,k=k,replicates=replicates,norm=norm,
conditions=conditions,pnr=pnr,nss=nss,v=v,lc=lc)
## Probability of differential expression
print("Computing probability of differential expression...")
prob.concounts <- probdeg(miMD$Ms,miMD$Ds,miMD$Mn,miMD$Dn)$prob
if (!is.null(assayData(input)$exprs))
todos <- rownames(as.matrix(assayData(input)$exprs))
else
todos <- rownames(as.matrix(assayData(input)$counts))
prob <- prob.concounts[todos]
names(prob) <- todos
## Results
resultat <- data.frame("level_1"=miMD$Level1,"level_2"=miMD$Level2,
"M"=miMD$Ms,"D"=miMD$Ds,"prob"=prob)
rownames(resultat) <- todos
# We change the name of the conditions to "name_mean"
colnames(resultat)[1] <-
paste(unlist(strsplit(miMD$comp," "))[1],"mean",sep="_")
colnames(resultat)[2] <-
paste(unlist(strsplit(miMD$comp," "))[3],"mean",sep="_")
resultat <- data.frame(resultat,"ranking"=ranking(resultat)$statistic)
if (!is.null(featureData(input)@data$Length))
resultat <- data.frame(resultat,
"Length"=as.numeric(as.character(featureData(input)@data[todos,
"Length"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$GC))
resultat <- data.frame(resultat,"GC"=as.numeric(as.character(
featureData(input)@data[todos,"GC"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Chromosome))
resultat <- data.frame(resultat,"Chrom"=as.character(
featureData(input)@data$Chromosome),"GeneStart"=as.numeric(
as.character(featureData(input)@data$GeneStart)),
"GeneEnd"=as.numeric(as.character(featureData(input)@data$GeneEnd)),
stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Biotype))
resultat <- data.frame(resultat,"Biotype"=as.character(
featureData(input)@data[todos,"Biotype"]),stringsAsFactors=FALSE)
#resultat[order(resultat[,5],decreasing=TRUE),]
Output(data=list(resultat),method=norm,k=miMD$k,lc=lc,factor=factor,v=v,
nss=nss,pnr=pnr,comparison=miMD$comp,replicates=replicates)
}
################################################################################
################################################################################
# noiseqbio.R
noiseqbio <- function (input,k=0.5,norm=c("rpkm","uqua","tmm","n"),nclust=15,
plot=FALSE,factor=NULL,conditions=NULL,lc=0,r=50,adj=1.5,a0per=0.9,filter=1,
depth=NULL,cv.cutoff=500,cpm=1) {
if (inherits(input,"eSet") == FALSE)
stop("ERROR: You must give an object generated by the readData ",
"function\n")
if (is.null(factor))
stop("ERROR: You must specify the factor to know which conditions ",
"you wish to compare. Please,give the argument 'factor'.\n")
if (min(table(input@phenoData@data[,factor])) < 2)
stop("ERROR: To run NOISeqBIO at least two replicates per condition ",
"are needed. Please, run NOISeq if there are not enough ",
"replicates in your experiment.\n")
norm <- match.arg(norm)
# Z-scores for signal and noise
cat("Computing Z values...\n")
miMD <- allMDbio(input,factor,k=k,norm=norm,conditions=conditions,lc=lc,
r=r,a0per=a0per,nclust=nclust,filter=filter,depth=depth,
cv.cutoff=cv.cutoff,cpm=cpm)
##### Probability of differential expression
cat("Computing probability of differential expression...\n")
## KDE estimators of f0 and f
desde <- min(c(miMD$Zs,miMD$Zn),na.rm=TRUE)
hasta <- max(c(miMD$Zs,miMD$Zn),na.rm=TRUE)
fdens <- density(miMD$Zs,adjust=adj,n=5000,from=desde,to=hasta,na.rm=TRUE)
f <- approxfun(fdens)
f0dens <- density(miMD$Zn,adjust=adj,n=5000,from=desde,to=hasta,na.rm=TRUE)
f0 <- approxfun(f0dens)
if (f0(0)/f(0) < 1)
print("WARNING: f0(0)/f(0) < 1 => FP with Z~0 will be detected.")
f0.f <- f0(miMD$Zs)/f(miMD$Zs)
## f0,f plot
if (plot) {
plot(f0dens,lwd=2,col=4,main=paste("r=",r,"; a0per=",a0per,sep=""),
xlab="theta")
lines(fdens,lwd=2,col=1)
legend("topleft",c("f","f0"),col=c(1,4),lwd=2)
}
## ESTIMATION of p0
p0 <- min(1/f0.f,na.rm=TRUE)
p0=min(p0,1)
cat(paste("p0 =",p0)); cat("\n")
## PROBABILITY of DIFFERENTIAL EXPRESSION
myprob <- 1 - p0*f0.f
if (min(myprob,na.rm=TRUE) < 0) {
print(summary(f0.f))
}
names(myprob) <- names(miMD$Zs)
cat("Probability\n")
print(summary(myprob))
if (!is.null(assayData(input)$exprs))
todos <- rownames(as.matrix(assayData(input)$exprs))
else
todos <- rownames(as.matrix(assayData(input)$counts))
myprob <- myprob[todos]
names(myprob) <- todos
## Results
resultat <- data.frame("level_1"=miMD$Level1,"level_2"=miMD$Level2,
"theta"=miMD$Zs,"prob"=myprob,"log2FC"=log2(miMD$Level1/miMD$Level2))
rownames(resultat) <- todos
colnames(resultat)[1] <-
paste(unlist(strsplit(miMD$comp," "))[1],"mean",sep="_")
colnames(resultat)[2] <-
paste(unlist(strsplit(miMD$comp," "))[3],"mean",sep="_")
# resultat <- data.frame(resultat,"ranking"=ranking(resultat)$statistic)
if (!is.null(featureData(input)@data$Length))
resultat <- data.frame(resultat,"length"=as.numeric(as.character(
featureData(input)@data[todos,"Length"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$GC))
resultat <- data.frame(resultat,"GC"=as.numeric(as.character(
featureData(input)@data[todos,"GC"])),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Chromosome))
resultat <- data.frame(resultat,"Chrom"=(as.character(
featureData(input)@data$Chromosome)),
"GeneStart"=as.numeric(as.character(featureData(
input)@data$GeneStart)),"GeneEnd"=as.numeric(as.character(
featureData(input)@data$GeneEnd)),stringsAsFactors=FALSE)
if (!is.null(featureData(input)@data$Biotype))
resultat <- data.frame(resultat,"Biotype"=as.character(
featureData(input)@data[todos,"Biotype"]),stringsAsFactors=FALSE)
Output(data=list(resultat),method=norm,k=miMD$k,lc=lc,factor=factor,
comparison=miMD$comp,replicates="biological",v=0,nss=0,pnr=0)
}
################################################################################
################################################################################
# normalization.R
noitmm <- function (datos,long=1000,lc=0,k=0,refColumn=1,logratioTrim=.3,
sumTrim=0.05,doWeighting=TRUE,Acutoff=-1e10) {
if (!is.null(ncol(long))) {
mynames <- long[,1]
long <- long[,2]
names(long) <- mynames
}
L <- (long/1000)^lc
datos <- datos/L
total <- colSums(as.matrix(datos))
datos0 <- sinceros(datos,k)
if (ncol(as.matrix(datos)) > 1) {
fk <- .calcNormFactors(as.matrix(datos),refColumn=refColumn,
method="TMM",logratioTrim=logratioTrim,sumTrim=sumTrim,
doWeighting=doWeighting,Acutoff=Acutoff)
fk <- fk * (total/mean(total))
datos.norm <- t(t(datos0)/fk)
}
else {
datos.norm <- datos0/L
}
na.omit(datos.norm)
}
noirpkm <- function (datos,long=1000,lc=1,k=0) {
if (!is.null(ncol(long))) {
mynames=long[,1]
long=long[,2]
names(long)=mynames
}
total <- colSums(as.matrix(datos))
datos0 <- sinceros(datos,k)
datos.norm <- (t(t(datos0)/total)*10^6)/((long/1000)^lc)
na.omit(datos.norm)
}
noiuqua <- function (datos,long=1000,lc=0,k=0) {
if (!is.null(ncol(long))) {
mynames=long[,1]
long=long[,2]
names(long)=mynames
}
L <- (long/1000)^lc
datos=datos/L
datos0 <- sinceros(datos,k)
if (ncol(as.matrix(datos)) > 1) {
sumatot <- rowSums(datos)
supertot <- sum(sumatot)
counts0 <- which(sumatot == 0)
if (length(counts0) > 0) {
datitos <- datos[-counts0,]
} else {
datitos <- datos
}
q3 <- apply(datitos,2,quantile,probs=0.75)
d <- q3*supertot/sum(q3)
datos.norm <- t(t(datos0)/d)*10^6
} else {
datos.norm <- datos0/L
}
na.omit(datos.norm)
}
## Taken from the edgeR package with minor modifications
.calcNormFactors <- function(object,method=c("TMM","quantile"),refColumn=NULL,
logratioTrim=.3,sumTrim=0.05,doWeighting=TRUE,Acutoff=-1e10,quantile=0.75) {
method <- match.arg(method)
if( is.matrix(object) ) {
if(is.null(refColumn))
refColumn <- 1
data <- object
libsize <- colSums(data)
}
else {
stop(".calcNormFactors() only operates on 'matrix' objects")
}
f <- switch(method,
TMM = apply(data,2,.calcFactorWeighted,ref=data[,refColumn],
logratioTrim=logratioTrim,sumTrim=sumTrim,doWeighting=doWeighting,
Acutoff=Acutoff),quantile=.calcFactorQuantile(data,libsize,
q=quantile))
f <- f/exp(mean(log(f)))
return(f)
}
.calcFactorQuantile <- function(data,lib.size,q=0.75) {
y <- t(t(data)/lib.size)
f <- apply(y,2,function(x) quantile(x,p=q))
f/exp(mean(log(f)))
}
.calcFactorWeighted <- function(obs,ref,logratioTrim=.3,sumTrim=0.05,
doWeighting=TRUE,Acutoff=-1e10) {
if( all(obs==ref) )
return(1)
obs <- as.numeric(obs)
ref <- as.numeric(ref)
nO <- sum(obs)
nR <- sum(ref)
# log ratio of expression,accounting for library size
logR <- log2((obs/nO)/(ref/nR))
absE <- (log2(obs/nO) + log2(ref/nR))/2 # absolute expression
v <- (nO-obs)/nO/obs + (nR-ref)/nR/ref # estimated asymptotic variance
# remove infinite values,cutoff based on A
fin <- is.finite(logR) & is.finite(absE) & (absE > Acutoff)
logR <- logR[fin]
absE <- absE[fin]
v <- v[fin]
# taken from the original mean() function
n <- sum(fin)
loL <- floor(n * logratioTrim) + 1
hiL <- n + 1 - loL
loS <- floor(n * sumTrim) + 1
hiS <- n + 1 - loS
#keep <- (rank(logR) %in% loL:hiL) & (rank(absE) %in% loS:hiS)
# a fix from leonardo ivan almonacid cardenas,since rank() can return
# non-integer values when there are a lot of ties
keep <- (rank(logR)>=loL & rank(logR)<=hiL) & (rank(absE)>=loS
& rank(absE)<=hiS)
if (doWeighting)
2^(sum(logR[keep]/v[keep],na.rm=TRUE) / sum(1/v[keep],na.rm=TRUE))
else
2^(mean(logR[keep],na.rm=TRUE))
}
################################################################################
################################################################################
# PCA.GENES.R
PCA.GENES <- function(X) {
n <- ncol(X)
p <- nrow(X)
offset <- apply(X,2,mean)
Xoff <- X-(cbind(matrix(1,p,1))%*%rbind(offset))
eigen <- eigen(Xoff%*%t(Xoff)/(p-1))
var <- cbind(eigen$values/sum(eigen$values),
cumsum(eigen$values/sum(eigen$values)))
loadings2 <- eigen$vectors
scores2 <- t(Xoff)%*%loadings2
normas2 <- sqrt(apply(scores2^2,2,sum))
scores1 <- loadings2%*%diag(normas2)
loadings1 <- scores2%*%diag(1/normas2)
output <- list(eigen,var,scores1,loadings1)
names(output) <- c("eigen","var.exp","scores","loadings")
output
}
################################################################################
################################################################################
# probdeg.R
probdeg <- function (Mg,Dg,Mn,Dn,prec=2) {
tot <- length(Mn) # number of points in noise distribution
gens <- names(Mg)
Mruido <- abs(round(Mn,prec))
Druido <- round(Dn,prec)
Mgen <- abs(round(Mg,prec))
Dgen <- round(Dg,prec)
MDgen <- na.omit(cbind(Mgen, Dgen))
MDunic <- unique(MDgen)
Nres <- apply(MDunic,1,n.menor,S1=Mruido,S2=Druido)
lugares <- apply(MDgen,1,busca,S=MDunic)
Nconj <- Nres[lugares]
names(Nconj) <- names(lugares)
laprob <- Nconj / tot
laprob <- laprob[gens]
names(laprob) <- gens
Nconj <- Nconj[gens]
names(Nconj) <- gens
laprob <- list("prob"=laprob,"numDE"=Nconj,"numNOISE"=tot)
laprob
}
################################################################################
################################################################################
# readData.R
readData <- function(data=NULL,factors=NULL,length=NULL,biotype=NULL,
chromosome=NULL,gc=NULL) {
if (is.null(data))
stop("Expression information must be provided to the readData function")
if (is.null(factors))
stop("Condition information must be provided to the readData funcion")
if (!is.null(length) && !is.vector(length) && !is.data.frame(length)
&& !is.matrix(length))
stop( "The length info should be a vector or a data.frame/matrix.")
if (!is.null(gc) && !is.vector(gc) && !is.data.frame(gc) && !is.matrix(gc))
stop( "The GC content info should be a vector or a data.frame/matrix.")
if (!is.null(chromosome) && ncol(chromosome) != 3)
stop( "The chromosome object should be a matrix or data.frame with ",
"3 columns: chromosome,start position and end position.")
if (!is.null(biotype) && !is.vector(biotype) && !is.data.frame(biotype)
&& !is.matrix(biotype))
stop( "The biotype info should be a vector or a data.frame/matrix.")
countData <- as.matrix(data)
rowNames <- rownames(countData)
if (nrow(factors) == ncol(countData)) {
rownames(factors) <- colnames(countData)
}
else {
stop ("Number of rows in factors must be equal to number of ",
"columns in data.\n")
}
pheno <- AnnotatedDataFrame(data=as.data.frame(factors))
input <- ExpressionSet(assayData=countData,phenoData=pheno)
if (!is.null(length))
input <- addData(data=input,length=length)
if (!is.null(gc))
input <- addData(data=input,gc=gc)
if (!is.null(biotype))
input <- addData(data=input,biotype=biotype)
if (!is.null(chromosome))
input <- addData(data=input,chromosome=chromosome)
input
}
addData <- function(data,length=NULL,biotype=NULL,chromosome=NULL,factors=NULL,
gc=NULL) {
if (!inherits(data,"eSet"))
stop("Error. You must give an eSet object.")
if (!is.null(length) && !is.vector(length) && !is.data.frame(length)
&& !is.matrix(length))
stop( "The length info should be a vector or a data.frame/matrix.")
if (!is.null(gc) && !is.vector(gc) && !is.data.frame(gc) && !is.matrix(gc))
stop( "The GC content info should be a vector or a data.frame/matrix.")
if (!is.null(biotype) && !is.vector(biotype) && !is.data.frame(biotype)
&& !is.matrix(biotype))
stop( "The biotype info should be a vector or a data.frame/matrix.")
if (!is.null(chromosome) && ncol(chromosome) != 3)
stop( "The chromosome object should be a matrix or data.frame with ",
"3 columns: chromosome,start position and end position.")
if (!is.null(assayData(data)$exprs))
rowNames <- rownames(assayData(data)$exprs)
else
rowNames <- rownames(assayData(data)$counts)
# If exists length
if (!is.null(length)) {
Length <- rep(NA,length(rowNames))
names(Length) <- rowNames
if (is.vector(length)) {
Length[rowNames] <- as.numeric(as.character(length[rowNames]))
}
else if (is.data.frame(length) || is.matrix(length)) {
if (ncol(length) == 2) {
# We assume that the feature names are in the first column and
# the length in the second
rownames(length) <- length[,1]
Length[rowNames] <- as.numeric(as.character(length[rowNames,2]))
}
else if (ncol(length) == 1) {
# We assume that the length are in the first column and the
# feature names in the rownames
Length[rowNames] <- as.numeric(as.character(
length[rowNames,1]))
}
else {
stop( "The length matrix/data.frame contains more columns ",
"than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,Length)
}
# If exists gc
if (!is.null(gc)) {
GC <- rep(NA,length(rowNames))
names(GC) <- rowNames
if (is.vector(gc)) {
GC[rowNames] <- as.numeric(as.character(gc[rowNames]))
}
else if (is.data.frame(gc) || is.matrix(gc)) {
if (ncol(gc) == 2) {
# We assume that the feature names are in the first column and
# the GC content in the second
rownames(gc) <- gc[,1]
GC[rowNames] <- as.numeric(as.character(gc[rowNames,2]))
}
else if (ncol(gc) == 1) {
# We assume that the GC contents are in the first column
# and the feature names in the rownames
GC[rowNames] <- as.numeric(as.character(gc[rowNames,1]))
}
else {
stop( "The GC matrix/data.frame contains more columns ",
"than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,GC)
}
# If exists biotype
if (!is.null(biotype)) {
Biotype <- rep(NA,length(rowNames))
names(Biotype) <- rowNames
if (is.vector(biotype)) {
Biotype[rowNames] <- as.character(biotype[rowNames])
}
else if (is.data.frame(biotype) || is.matrix(biotype)) {
if (ncol(biotype) == 2) {
# We assume that the feature names are in the first column and
# the biotypes in the second
rownames(biotype) <- biotype[,1]
Biotype[rowNames] <- as.character( biotype[rowNames,2])
}
else if (ncol(biotype) == 1) {
# We assume that the biotypes are in the first column and
# the feature names in the rownames
Biotype[rowNames] <- as.character( biotype[rowNames,1] )
}
else {
stop("The biotype matrix/data.frame contains more ",
"columns than expected.")
}
}
featureData(data)@data <- cbind(featureData(data)@data,Biotype)
featureData(data)@data$Biotype <-
as.character(featureData(data)@data$Biotype)
}
# If exists chromosome
if (!is.null(chromosome)) {
Chromosome <- GeneStart <- GeneEnd <- rep(NA,length(rowNames))
names(Chromosome) <- names(GeneStart) <- names(GeneEnd) <- rowNames
Chromosome[rowNames] <- as.character(chromosome[rowNames,1])
GeneStart[rowNames] <- as.numeric(as.character(chromosome[rowNames,2]))
GeneEnd[rowNames] <- as.numeric(as.character(chromosome[rowNames,3]))
featureData(data)@data <- cbind(featureData(data)@data,Chromosome,
GeneStart,GeneEnd)
}
# If exists new factors
if (!is.null(factors))
phenoData(data)@data <- cbind(phenoData(data)@data,factors)
data
}
################################################################################
################################################################################
# saturation.plot.R
#### SATURATION PLOTS
## Data for saturation plot (with or without biotypes)
saturation.dat <- function (input,k=0,biotypes=NULL,ndepth=6) {
if (inherits(input,"eSet") == FALSE)
stop("Error. You must give an eSet object\n")
if (!is.null(assayData(input)$exprs))
datos <- assayData(input)$exprs
else
datos <- assayData(input)$counts
if (!is.null(featureData(input)$Biotype))
infobio <- featureData(input)$Biotype
else
infobio <- NULL
nsam <- NCOL(datos)
if (!is.null(infobio)) {
if(is.null(biotypes)) {
biotypes <- unique(infobio)
names(biotypes) <- biotypes
}
} else { biotypes=NULL }
satura <- vector("list",length=length(biotypes)+1)
names(satura) <- c("global",names(biotypes))
ndepth1 <- ceiling(ndepth/2)
datos <- round(datos,0)
datos0 <- datos + 0.2
# Random subsamples for each sample
submuestras <- seq.depth <- vector("list",length=nsam)
names(submuestras) <- names(seq.depth) <- colnames(datos)
for (n in seq_len(nsam)) { # simulating subsamples for each sample
total <- sum(datos[,n]) # total counts in sample n
# simulation for each depth and real depth
varias <- vector("list",length=ndepth+1)
for (i in seq_len(ndepth1)) { # simulating depths < real depth
muestra <- rmultinom(10,size=round(total/(ndepth1+1),0),
prob=datos[,n])
if (i == 1)
varias[[i]] <- muestra
else
varias[[i]] <- varias[[i-1]] + muestra
}
varias[[ndepth1+1]] <- as.matrix(datos[,n])
for (i in (ndepth1+2):(ndepth+1)) { # simulating depths < real depth
muestra <- rmultinom(10,size=round(total/(ndepth1+1),0),
prob=datos0[,n])
if (i == ndepth1+2)
varias[[i]] <- matrix(varias[[i-1]],ncol=10,
nrow=nrow(varias[[i-1]])) + muestra
else
varias[[i]] <- varias[[i-1]] + muestra
}
submuestras[[n]] <- varias
seq.depth[[n]] <- c(round(total/(ndepth1+1),0)*seq_len(ndepth1),total,
round(total/(ndepth1+1),0)*((ndepth1+2):(ndepth+1)))
}
# Global saturation
satura[[1]] <- vector("list",length=nsam)
names(satura[[1]]) <- colnames(datos)
for (n in seq_len(nsam)) { # for each sample
satura[[1]][[n]] <- vapply(submuestras[[n]],function(x) {
mean(apply(x,2,noceros,k=k))
},numeric(1))
}
# Per biotypes
if (!is.null(infobio)) { # if biotypes available
biog <- lapply(biotypes,function(x) {
which(is.element(infobio,x))
})
names(biog) <- names(biotypes)
for (j in 2:length(satura)) { # for each biotype
satura[[j]] <- vector("list",length=nsam)
names(satura[[j]]) <- colnames(datos)
for (n in seq_len(nsam)) { # for each sample
conbio <- lapply(submuestras[[n]],function(x) {
as.matrix(x[biog[[j-1]],])
})
satura[[j]][[n]] <- vapply(conbio,function(x) {
mean(apply(x,2,noceros,k=k))
},numeric(1))
}
}
}
else
biog <- NULL
# computing detection increasing per million reads
newdet <- vector("list",length=1+length(biotypes))
names(newdet) <- c("global",names(biotypes))
for (j in seq_len(length(newdet))) {
newdet[[j]] <- vector("list",length=nsam)
names(newdet[[j]]) <- colnames(datos)
for (n in seq_len(nsam)) {
puntos <- data.frame("x"=seq.depth[[n]],"y"=satura[[j]][[n]])
pendi <- NULL
for(i in 2:nrow(puntos))
pendi <- c(pendi,
(puntos$y[i]-puntos$y[i-1])/(puntos$x[i]-puntos$x[i-1]))
newdet[[j]][[n]] <- c(NA,pendi*1000000)
}
}
bionum <- c(NROW(datos),vapply(biog,length,integer(1)))
names(bionum) <- c("global",names(biog))
# Results at real sequencing depth
real <- vector("list",length=length(satura))
names(real) <- names(satura)
realdepth <- vapply(seq.depth,function (x) x[ndepth1+1],numeric(1))/10^6
for (i in seq_len(length(real))) {
real[[i]] <- data.frame("depth"=realdepth,
"detec"=vapply(satura[[i]],function (x) x[ndepth1+1],numeric(1)))
rownames(real[[i]]) <- colnames(datos)
}
# Results
satura <- list("saturation"=satura,"bionum"=bionum,"depth"=seq.depth,
"newdet"=newdet,"real"=real)
satura
}
################################################################################
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.