I am running R 4.1.0 with MixOmics 6.17.28
When follow the Bookdown Document ( Chapter 3 Principal Component Analysis (PCA) | mixOmics vignette ) on section3 all the PlotIndiv graphs only display colored dots not the variable name like in your examples? None of the PlotIndiv show labels but i chose one one image. Also note the Leged does not increase in numerical order like in the example.
I cut and pasted the code from the Bookdown Document so I would not make a mistake
Also wanted to note when i ran the same code with the previous version I had mixOmics 6.16.3 the labels printed fine. Since i had issues with the legen not displaying in incremental numerical order I updated to the newer MixOmics 6.17.28. Is there a way i can force the labels on the newer version or do i need to revert bacK to 6.16.3
Also for some reason in MixOmics 6.17.28 when I run the sPCA example the values for the axis display as 0 .
When i try to get the values for PC1 and PC2 in the sPCA example above i get for MyResult.spca :
sparse PCA with 3 principal components.
which does not match the Bookdown document values of PC1=23% and PC2 17%
You should post your code, just to be sure
I copy and pasted the code from the Bookdown document.
Can you paste the code you have… even if its from the bookdown document…
####---------------------Section 3 Bookdown------------PCA& sPCA------------
# combined data: liver.toxicity is a list of 4 df: gene, clinic, treatment, gene.ID
data(liver.toxicity)
names(liver.toxicity)
#[1] "gene" "clinic" "treatment" "gene.ID"
#Separate liver.toxicity into indiv df's'
genedf <- liver.toxicity$gene # no zeroes data looks centered
treatdf<- liver.toxicity$treatment
clinicdf<- liver.toxicity$clinic
gene.IDdf<- liver.toxicity$gene.ID
#Note: if the names of the samples are not meaningful at this stage, they can be replaced
#by symbols (ind.names=TRUE) test if setting ind.names=F forces the names to show
#PCA
MyResult.pca <- pca(genedf) # 1 Run the method
plotIndiv(MyResult.pca) # 2 Plot the samples orig form plots circles
plotVar(MyResult.pca, cutoff = 0.8) # 3 Plot the variables
##Customize PLots-----
#note Dose.Group is a column header in the dataframe liver.toxicity$treatments
#orig using imbedded data name
plotIndiv(MyResult.pca, group = liver.toxicity$treatment$Dose.Group,
legend = TRUE)
#without the nested file name
#plots exaclty same as above command
plotIndiv(MyResult.pca, group = treatdf$Dose.Group,
legend = TRUE)
#cutomize legends, Titles
#Additionally, two factors can be displayed using both colours (argument group) and symbols (argument pch)
#here exposure is the Time.Group column
plotIndiv(MyResult.pca, ind.names = FALSE,
group = liver.toxicity$treatment$Dose.Group,
pch = as.factor(liver.toxicity$treatment$Time.Group),
legend = TRUE, title = 'Liver toxicity: genes, PCA comp 1 - 2',
legend.title = 'Dose', legend.title.pch = 'Exposure')
# not using the nested file names # same graph as above
plotIndiv(MyResult.pca, ind.names = FALSE,
group = treatdf$Dose.Group,
pch = as.factor(treatdf$Time.Group),
legend = TRUE, title = 'Liver toxicity: genes, PCA comp 1 - 2',
legend.title = 'Dose', legend.title.pch = 'Exposure')
#PCA with 3 components
MyResult.pca2 <- pca(genedf, ncomp = 3)
plotIndiv(MyResult.pca2, comp = c(1,3), legend = TRUE,
group = liver.toxicity$treatment$Time.Group,
title = 'Multidrug transporter, PCA comp 1 - 3')
#Note the 3rd component in Y axis highlights an exposure effect
# amount of variance explained & # of components
plot(MyResult.pca2)
# first 3 components explain approx 65% variance see below exace 62.7
MyResult.pca2
# Eigenvalues for the first 3 principal components, see object$sdev^2:
# PC1 PC2 PC3
# 17.971416 9.079234 4.567709
#
# Proportion of explained variance for the first 3 principal components, see object$prop_expl_var:
# PC1 PC2 PC3
# 0.35684128 0.18027769 0.09069665
#
# Cumulative proportion of explained variance for the first 3 principal components, see object$cum.var:
# PC1 PC2 PC3
# 0.3568413 0.5371190 0.6278156
#
# Other available components:
#
# loading vectors: see object$rotation
# Other functions:
# plotIndiv, plot, plotVar, selectVar, biplot
#other plots
# The loading weights are represented in decreasing order
# from bottom to top in plotLoadings. Their absolute value
# indicates the importance of each variable to define each
# PC, as represented by the length of each bar
# a customized example to only show the top 100 genes
# and their gene name
plotLoadings(MyResult.pca, ndisplay = 100,
name.var = liver.toxicity$gene.ID[, "geneBank"],
size.name = rel(0.3))
#variable selection with sparse PCS
#Apply PCA and identify the key variables that contribute
#to the explanation of most variance in the data set
# The user needs to provide the number of variables to
# select on each PC. Here, for example, we ask to select
# the top 15 genes contributing to the definition of
# PC1, the top 10 genes contributing to PC2 and the
# top 5 genes for PC3 (keepX=c(15,10,5)).
X <-genedf
MyResult.spca <- spca(X, ncomp = 3, keepX = c(15,10,5)) # 1 Run the method
plotIndiv(MyResult.spca, group = liver.toxicity$treatment$Dose.Group, # 2 Plot the samples
pch = as.factor(liver.toxicity$treatment$Time.Group),
legend = TRUE, title = 'Liver toxicity: genes, sPCA comp 1 - 2',
legend.title = 'Dose', legend.title.pch = 'Exposure')
# since axi PCA values above show O use below to get actual values
MyResult.spca
# sparse PCA with 3 principal components.
# Input data X of dimensions: 64 3116
# Number of selected variables on each prinicipal components:
# PC1 PC2 PC3
# 15 10 5
# Proportion of adjusted explained variance for the first 3 principal components, see object$prop_expl_var:
# PC1 PC2 PC3
# 0.003856574 0.002213243 0.001108384
#
# Other available components:
#
# loading vectors: see object$rotation
# Other functions:
#
# tune.spca, plotIndiv, plot, plotVar, selectVar, biplot
plotVar(MyResult.spca, cex = 1) # 3 Plot the variables
#Selected variables can be identified on each component
#with the selectVar function. Here the coefficient
#values are extracted,see ?selectVar:
selectVar(MyResult.spca, comp = 1)$value
# Those values correspond to the loading weights that
# are used to define each component. A large absolute
# value indicates the importance of the variable in this
# PC. Selected variables are ranked from the most
# important (top) to the least important
# value.var
# A_43_P20281 -0.39077443
# A_43_P16829 -0.38898291
# A_43_P21269 -0.37452039
# A_43_P20475 -0.32482960
# A_43_P20891 -0.31740002
# A_43_P14037 -0.27681845
# A_42_P751969 -0.26140533
# A_43_P15845 -0.22392912
# A_42_P814129 -0.18838954
# A_42_P680505 -0.18672610
# A_43_P21483 -0.16202222
# A_43_P21243 -0.13259471
# A_43_P22469 -0.12493156
# A_43_P23061 -0.12255308
# A_43_P11409 -0.09768656
plotLoadings(MyResult.spca) # comp 1
#for component 2
selectVar(MyResult.spca, comp=2)$value
plotLoadings(MyResult.spca, comp = 2)
#Tuning PCA
tune.pca(X)# does nothing on MixOmix vs 6.17.28
####---------------------Section 4 Bookdown-------PLS-DA------------
#I am analysing a single data set (e.g. transcriptomics data) and I would like to classify
#my samples into known groups and predict the class of new samples. In addition, I am interested
#in identifying the key variables that drive such discrimination.
data(srbct) #the expression levels of 2,308 genes measured on 63 samples
names(srbct)
#[1] "gene" "class" "gene.name"
X <- srbct$gene #a data frame with 63 rows and 2308 columns. The expression levels of 2,308 genes in 63 subjects.
Y <- srbct$class #a class vector containing the class tumour of each individual (4 classes in total)
Z <- srbct$gene.name # a data frame with 2,308 rows and 2 columns containing further information on the genes.
summary(Y) ## class summary
# EWS BL NB RMS
# 23 8 12 20
dim(X) ## number of samples and features
#[1] 63 2308
length(Y) ## length of class memebrship factor = number of samples
#[1] 63
#sLPCA-DA
MyResult.splsda <- splsda(X, Y, keepX = c(50,50)) # 1 Run the method
plotIndiv(MyResult.splsda) # 2 Plot the samples (coloured by classes automatically)
plotVar(MyResult.splsda) # 3 Plot the variables
#select var Comp1
selectVar(MyResult.splsda, comp=1)$name # Selected variables on component 1
#PLS-DA without variable selection
MyResult.plsda <- plsda(X,Y) # 1 Run the method
plotIndiv(MyResult.plsda) # 2 Plot the samples
plotVar(MyResult.plsda, cutoff = 0.7) # 3 Plot the variables
#Customizing Sample Plots
#The sample plots can be improved in various ways. First, if the names of the samples are
#not meaningful at this stage, they can be replaced by symbols (ind.names=TRUE).
#Confidence ellipses can be plotted for each sample (ellipse = TRUE, confidence level set to 95%
#by default, see the argument ellipse.level), Additionally, a star plot displays arrows from eac
#group centroid towards each individual sample (star = TRUE)
plotIndiv(MyResult.splsda, ind.names = FALSE, legend=TRUE,
ellipse = TRUE, star = TRUE, title = 'sPLS-DA on SRBCT',
X.label = 'PLS-DA 1', Y.label = 'PLS-DA 2')
#Predictions(classification)
background <- background.predict(MyResult.splsda, comp.predicted=2,
dist = "max.dist")
plotIndiv(MyResult.splsda, comp = 1:2, group = srbct$class,
ind.names = FALSE, title = "Maximum distance",
legend = TRUE, background = background)
#Variable selection
MyResult.splsda2 <- splsda(X,Y, ncomp=3, keepX=c(15,10,5))
selectVar(MyResult.splsda2, comp=1)$value
# value.var
# g123 0.53516982
# g846 0.41271455
# g335 0.30309695
# g1606 0.30194141
# g836 0.29365241
# g783 0.26329876
# g758 0.25826903
# g1386 0.23702577
# g1158 0.15283961
# g585 0.13838913
# g589 0.12738682
# g1387 0.12202390
# g1884 0.08458869
# g1295 0.03150351
# g1036 0.00224886
plotLoadings(MyResult.splsda2, contrib = 'max', method = 'mean')
#tunin Parameters and numerical output
MyResult.plsda2 <- plsda(X,Y, ncomp=10)
set.seed(30) # for reproducbility in this vignette, otherwise increase nrepeat
MyPerf.plsda <- perf(MyResult.plsda2, validation = "Mfold", folds = 3,
progressBar = FALSE, nrepeat = 10) # we suggest nrepeat = 50
plot(MyPerf.plsda, col = color.mixo(5:7), sd = TRUE, legend.position = "horizontal")
type or paste code here
Problem 1:
You need to change ind.names=FALSE to ind.names=TRUE
The issue is with the version of MixOmics. When i use version MixOmics 6.17.28 even with indiv.names=TRUE all i see is the colored dots. When i reverted to MixOmicx 6.16.3 the labeles worked fine.
hi @Ivette,
Thank you for reporting the bug, it happens when Bioconductor upgrades (every 6 months) and the development version of the package is then pushed on Bioconductor. Similarly we then update the vignette accordingly. We’ll get to that issue soon with @aljabadi. I suggest you stick to 6.16 for now.
Kim-Anh
Hi @Ivette, thank you for reporting this. We highly recommend using the release version of mixOmics, especially when such issues arise in the development version. Often such issues are caused by unstable versions of dependencies.