Nothing
## ----style, echo = FALSE, results = 'asis'------------------------------------
BiocStyle::markdown()
## ----env, include=FALSE, echo=FALSE, cache=FALSE------------------------------
library("knitr")
opts_chunk$set(stop_on_error = 1L)
suppressPackageStartupMessages(library("MSnbase"))
suppressWarnings(suppressPackageStartupMessages(library("pRoloc")))
suppressPackageStartupMessages(library("pRolocdata"))
suppressPackageStartupMessages(library("class"))
set.seed(1)
## ----libraries----------------------------------------------------------------
library("MSnbase")
library("pRoloc")
library("pRolocdata")
## ----setcols------------------------------------------------------------------
setStockcol(NULL) ## reset first
setStockcol(paste0(getStockcol(), 70))
## ----readCsvData0-------------------------------------------------------------
## The original data for replicate 1, available
## from the pRolocdata package
f0 <- dir(system.file("extdata", package = "pRolocdata"),
full.names = TRUE,
pattern = "pr800866n_si_004-rep1.csv")
csv <- read.csv(f0)
## ----showOrgCsv---------------------------------------------------------------
head(csv, n=3)
## ----readCsvData1, tidy = FALSE-----------------------------------------------
## The quantitation data, from the original data
f1 <- dir(system.file("extdata", package = "pRolocdata"),
full.names = TRUE, pattern = "exprsFile.csv")
exprsCsv <- read.csv(f1)
## Feature meta-data, from the original data
f2 <- dir(system.file("extdata", package = "pRolocdata"),
full.names = TRUE, pattern = "fdataFile.csv")
fdataCsv <- read.csv(f2)
## Sample meta-data, a new file
f3 <- dir(system.file("extdata", package = "pRolocdata"),
full.names = TRUE, pattern = "pdataFile.csv")
pdataCsv <- read.csv(f3)
## ----showExprsFile------------------------------------------------------------
head(exprsCsv, n=3)
## ----showFdFile---------------------------------------------------------------
head(fdataCsv, n=3)
## ----showPdFile---------------------------------------------------------------
pdataCsv
## ----makeMSnSet---------------------------------------------------------------
tan2009r1 <- readMSnSet(exprsFile = f1,
featureDataFile = f2,
phenoDataFile = f3,
sep = ",")
tan2009r1
## ----readMSnSet2--------------------------------------------------------------
ecol <- paste("area", 114:117, sep = ".")
fname <- "Protein.ID"
eset <- readMSnSet2(f0, ecol, fname)
eset
## ----ecols--------------------------------------------------------------------
getEcols(f0, ",")
grepEcols(f0, "area", ",")
e <- grepEcols(f0, "area", ",")
readMSnSet2(f0, e)
## ----showSubset---------------------------------------------------------------
smallTan <- tan2009r1[1:5, 1:2]
dim(smallTan)
exprs(smallTan)
## ----rmtan, echo=FALSE--------------------------------------------------------
## remove manullay constructred data
rm(tan2009r1)
data(tan2009r1)
## ----loadTan1-----------------------------------------------------------------
data(tan2009r1)
## ----lookAtTan----------------------------------------------------------------
getMarkers(tan2009r1, fcol = "markers.orig")
getMarkers(tan2009r1, fcol = "PLSDA")
## ----markers------------------------------------------------------------------
pRolocmarkers()
head(pRolocmarkers("dmel"))
table(pRolocmarkers("dmel"))
## ----realtiveQuants-----------------------------------------------------------
summary(rowSums(exprs(tan2009r1)))
## ----norm, echo=TRUE, message=FALSE, cache=TRUE-------------------------------
## create a small illustrative test data
data(itraqdata)
itraqdata <- quantify(itraqdata, method = "trap",
reporters = iTRAQ4)
## the quantification data
head(exprs(itraqdata), n = 3)
summary(rowSums(exprs(itraqdata)))
## normalising to the sum of feature intensitites
itraqnorm <- normalise(itraqdata, method = "sum")
processingData(itraqnorm)
head(exprs(itraqnorm), n = 3)
summary(rowSums(exprs(itraqnorm)))
## ----featurenames-------------------------------------------------------------
head(featureNames(tan2009r1))
## ----showplotdist, echo=TRUE, eval=FALSE--------------------------------------
# ## indices of the mito markers
# j <- which(fData(tan2009r1)$markers.orig == "mitochondrion")
# ## indices of all proteins assigned to the mito
# i <- which(fData(tan2009r1)$PLSDA == "mitochondrion")
# plotDist(tan2009r1[i, ],
# markers = featureNames(tan2009r1)[j])
## ----plotdist1, echo=FALSE, fig.cap="Distribution of protein intensities along the fractions of the separation gradient for 4 organelles: mitochondrion (red), ER/Golgi (blue, ER markers and green, Golgi markers) and plasma membrane (purple)."----
par(mfrow = c(2,2), mar = c(4, 4, 2, 0.1))
cls <- getStockcol()[1:4]
plotDist(tan2009r1[which(fData(tan2009r1)$PLSDA == "mitochondrion"), ],
markers = featureNames(tan2009r1)
[which(fData(tan2009r1)$markers.orig == "mitochondrion")],
mcol = cls[1])
title(main = "mitochondrion")
plotDist(tan2009r1[which(fData(tan2009r1)$PLSDA == "ER/Golgi"), ],
markers = featureNames(tan2009r1)
[which(fData(tan2009r1)$markers.orig == "ER")],
mcol = cls[2])
title(main = "ER")
plotDist(tan2009r1[which(fData(tan2009r1)$PLSDA == "ER/Golgi"), ],
markers = featureNames(tan2009r1)
[which(fData(tan2009r1)$markers.orig == "Golgi")],
mcol = cls[3])
title(main = "Golgi")
plotDist(tan2009r1[which(fData(tan2009r1)$PLSDA == "PM"), ],
markers = featureNames(tan2009r1)
[which(fData(tan2009r1)$markers.orig == "PM")],
mcol = cls[4])
title(main = "PM")
## ----dendro, fig.cap="Hierarchical clustering. Average distance between organelle classes."----
mrkHClust(tan2009r1)
## ----plotavrprof, fig.cap="Average organelle class profiles. Protein intensity indicated by colour. Organelle classes ordered by hierarchical clustering"----
## histogram
hc <- mrkHClust(tan2009r1, plot=FALSE)
## order of markers according to histogram
mm <- getMarkerClasses(tan2009r1)
m_order <- levels(factor(mm))[order.dendrogram(hc)]
## average marker profile
fmat <- mrkConsProfiles(tan2009r1)
plotConsProfiles(fmat, order=m_order)
## ----plot2d, echo=TRUE, fig.asp=1, fig.cap="PCA plot. Representation of the proteins of `tan2009r1` after reduction of the 4 reporter quantitation data to 2 principal components."----
plot2D(tan2009r1, fcol = "markers")
addLegend(tan2009r1, fcol = "markers", cex = .7,
where = "bottomright", ncol = 2)
## ----plot2dorg, echo=TRUE, fig.asp=1, fig.cap="PCA plot. Reduced set of markers for the `tan2009r1` data projected onto 2 principal components."----
plot2D(tan2009r1, fcol = "markers.orig")
addLegend(tan2009r1, fcol = "markers.orig", where = "bottomright")
## ----plot3Danim, echo=FALSE, eval=FALSE---------------------------------------
# ## here's how the animation below was created
# data(hyperLOPIT2015)
# suppressPackageStartupMessages(library("rgl"))
# plot3D(hyperLOPIT2015)
# movie3d(spin3d(c(1, 1, 1), rpm = 10), duration = 5,
# clean = FALSE,
# dir = "~/dev/00_github/pRoloc/vignettes/Figures/")
## ----plot3D, eval = FALSE-----------------------------------------------------
# plot3D(tan2009r1)
## ----scree0, echo=FALSE-------------------------------------------------------
pcvar <- plot2D(tan2009r1, method = "scree", plot = FALSE)
pcvar <- round(pcvar, 2)
## ----scree, fig.cap="Percentage of variance explained."-----------------------
plot2D(tan2009r1, method = "scree")
## ----tsne, eval=FALSE---------------------------------------------------------
# perps <- sort(c(30, seq(5, 50, 15)))
# data(HEK293T2011)
# par(mfrow = c(2, 3))
# plot2D(HEK293T2011, main = "PCA")
# sapply(perps,
# function(perp) {
# plot2D(HEK293T2011, method = "t-SNE",
# methargs = list(perplexity = perp))
# title(main = paste("t-SNE, perplexity", perp))
# })
## ----foi0---------------------------------------------------------------------
foi1 <- FeaturesOfInterest(description = "Feats of interest 1",
fnames = featureNames(tan2009r1[1:10]))
description(foi1)
foi(foi1)
## -----------------------------------------------------------------------------
foi2 <- FeaturesOfInterest(description = "Feats of interest 2",
fnames = featureNames(tan2009r1[880:888]))
foic <- FoICollection(list(foi1, foi2))
foic
## ----plotfoi, fig.asp=1, fig.cap="Adding features of interest on a PCA plot."----
plot2D(tan2009r1, fcol = "PLSDA")
addLegend(tan2009r1, fcol = "PLSDA",
where = "bottomright",
cex = .7)
highlightOnPlot(tan2009r1, foi1,
col = "black", lwd = 2)
highlightOnPlot(tan2009r1, foi2,
col = "purple", lwd = 2)
legend("topright", c("FoI 1", "FoI 2"),
bty = "n", col = c("black", "purple"),
pch = 1)
## ----guiinstall, eval = FALSE-------------------------------------------------
# library("BiocManager")
# BiocManager::install("pRolocGUI")
# library("pRolocGUI")
# pRolocVis(tan2009r1)
## ----qsep---------------------------------------------------------------------
library("pRolocdata")
data(hyperLOPIT2015)
## Create the object and get a summary
hlq <- QSep(hyperLOPIT2015)
hlq
levelPlot(hlq)
plot(hlq)
## ----plot2dnull, echo=TRUE, fig.cap="Plain PCA representation of the `tan2009r1` data."----
plot2D(tan2009r1, fcol = NULL)
## ----plotKmeans, echo=TRUE, fig.asp=1, fig.cap="k-means clustering on the `tan2009r1` data."----
kcl <- MLearn( ~ ., tan2009r1, kmeansI, centers=5)
plot(kcl, exprs(tan2009r1))
## ----plotHclust, echo=TRUE, tidy = FALSE, fig.asp=1, fig.cap="Hierarchical clustering on the `tan2009r1` data."----
hcl <- MLearn( ~ ., tan2009r1,
hclustI(distFun = dist,
cutParm = list(k = 5)))
plot(hcl, labels = FALSE)
## ----plotPam, echo=TRUE, fig.asp=1, fig.cap="Partitioning around medoids on the `tan2009r1` data."----
pcl <- MLearn( ~ ., tan2009r1, pamI(dist), k = 5)
plot(pcl, data = exprs(tan2009r1))
## ----svmParamOptim, eval = FALSE, message = FALSE, tidy=FALSE-----------------
# params <- svmOptimisation(tan2009r1, fcol = "markers.orig",
# times = 100, xval = 5,
# verbose = FALSE)
## ----loadParams, echo = TRUE--------------------------------------------------
fn <- dir(system.file("extdata", package = "pRoloc"),
full.names = TRUE, pattern = "params.rda")
load(fn)
## ----showParams---------------------------------------------------------------
params
## ----params1, echo=TRUE, fig.cap="Assessing parameter optimisation. Here, we see the respective distributions of the 10 macro F1 scores for the best cost/sigma parameter pairs. See also the output of *f1Count* in relation to this plot."----
plot(params)
## ----params2, echo=TRUE, fig.cap="Assessing parameter optimisation. Visualisation of the averaged macro F1 scores, for the full range of parameter values."----
levelPlot(params)
## ----f1count------------------------------------------------------------------
f1Count(params)
getParams(params)
## ----svmRes0, warning=FALSE, eval = FALSE, tidy = FALSE-----------------------
# ## manual setting of parameters
# svmres <- svmClassification(tan2009r1, fcol = "markers.orig",
# sigma = 1, cost = 1)
## ----svmRes, warning=FALSE, tidy = FALSE--------------------------------------
## using default best parameters
svmres <- svmClassification(tan2009r1, fcol = "markers.orig",
assessRes = params)
processingData(svmres)
tail(fvarLabels(svmres), 4)
## ----getPredictions-----------------------------------------------------------
p1 <- getPredictions(svmres, fcol = "svm")
p1 <- fData(p1)$svm.pred
minprob <- median(fData(svmres)$svm.scores)
p2 <- getPredictions(svmres, fcol = "svm", t = minprob)
p2 <- fData(p2)$svm.pred
table(p1, p2)
## ----predscores, fig.cap="Organelle-specific SVM score distributions."--------
boxplot(svm.scores ~ svm, data = fData(svmres),
ylab = "SVM scores")
abline(h = minprob, lwd = 2, lty = 2)
## ----medscores2, tidy = FALSE-------------------------------------------------
ts <- orgQuants(svmres, fcol = "svm", t = .5)
## ----medscorepreds, tidy = FALSE----------------------------------------------
getPredictions(svmres, fcol = "svm", t = ts)
## ----svmres, fig.asp=1, fig.cap="Representation of the svm prediction on the `tan2009r1` data set. The svm scores have been used to set the point size (`cex` argument; the scores have been transformed to emphasise the extremes). Different symbols (`fpch`) are used to differentiate markers and new assignments."----
ptsze <- exp(fData(svmres)$svm.scores) - 1
plot2D(svmres, fcol = "svm", fpch = "markers.orig",
cex = ptsze)
addLegend(svmres, fcol = "svm",
where = "bottomright",
cex = .5)
## ----runPhenoDisco, eval=FALSE, warning=FALSE---------------------------------
# pdres <- phenoDisco(tan2009r1, GS = 10,
# times = 100, fcol = "PLSDA")
## ----loadpdres, echo=TRUE-----------------------------------------------------
fn <- dir(system.file("extdata", package = "pRoloc"),
full.names = TRUE, pattern = "pdres.rda")
load(fn)
## ----phenoDiscoFvar-----------------------------------------------------------
processingData(pdres)
tail(fvarLabels(pdres), 3)
## ----pdres, echo=TRUE, fig.asp=1, fig.cap="Representation of the phenoDisco prediction and cluster discovery results."----
plot2D(pdres, fcol = "pd")
addLegend(pdres, fcol = "pd", ncol = 2,
where = "bottomright",
cex = .5)
## ----pdmarkers----------------------------------------------------------------
getMarkers(tan2009r1, fcol = "pd.markers")
## ----weights, eval = TRUE, echo = TRUE----------------------------------------
w <- classWeights(tan2009r1, fcol = "pd.markers")
w
## ----pdsvmParams, eval = FALSE, tidy = FALSE----------------------------------
# params2 <- svmOptimisation(tan2009r1, fcol = "pd.markers",
# times = 10, xval = 5,
# class.weights = w,
# verbose = FALSE)
## ----loadParams2, echo = TRUE-------------------------------------------------
fn <- dir(system.file("extdata", package = "pRoloc"),
full.names = TRUE, pattern = "params2.rda")
load(fn)
## ----pdsvm, cache = TRUE, message = FALSE, warning = FALSE, tidy = FALSE------
tan2009r1 <- svmClassification(tan2009r1, params2,
class.weights = w,
fcol = "pd.markers")
## ----pdres2, fig.asp=1, fig.cap="Classification results. The results of the second round of classification, using the augmented set of markers obtained using *phenoDisco* as detailed in [@Breckels2013] and a weighted svm classifier."----
ptsze <- exp(fData(tan2009r1)$svm.scores) - 1
plot2D(tan2009r1, fcol = "svm", cex = ptsze)
addLegend(tan2009r1, fcol = "svm",
where = "bottomright",
ncol = 2, cex = .5)
## ----sessioninfo, echo=FALSE--------------------------------------------------
sessionInfo()
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.