HOW to define the "parLabVc" indices in ropls package.

Question: HOW to define the "parLabVc" indices in ropls package.

5 months ago by

Hi, I am using package"ropls" to generate PCA plot. And I want to change the labels of observations on the plot, which is defaulted as the row name of m data set.

But, I do not know how to define the labels ("indices of "x") that I want to display in my PCA plot.

following the argument in the manual:

"parLabVc, Optional character vector for the labels of observations on the plot; default is NA [ie row names of ’x’, if available, or indices of ’x’, otherwise, will be used]"

ropls R

ADD COMMENT • link •

modified 5 months ago by etienne.thevenot • 0 • written 5 months ago by yuan.guo001 • 0

5 months ago by

etienne.thevenot • 0

France

etienne.thevenot • 0 wrote:

Hi,

Below is an example of labels ("s1", "s2", ..., "sN"), where N is the total number of samples. You can use any character vector of length N. Does it answer your question?

Etienne.

> library(ropls)
> data(sacurine)

> pcaModel <- opls(sacurine[["dataMatrix"]])
> plot(pcaModel, parLabVc = paste0("s", 1:nrow(sacurine[["dataMatrix"]])), typeVc = "x-score")

ADD COMMENT • link written 5 months ago by etienne.thevenot • 0

Hi Etienne,

thank you for your answer. My observations belong to 5 groups. Can I assign each group with symbols of "circles", "triangles" and so on, respectively? That will be more clean that just like"S1,S2,S3,S4......." on the plot

Thank you for your help.

Yuan

ADD REPLY • link written 5 months ago by yuan.guo001 • 0

You cannot use symbols (but special characters work). A better option would be to use colors according to your classes (use a factor in the parColFcVn argument):

> library(ropls)
> data(sacurine)
> pcaModel <- opls(sacurine[["dataMatrix"]])
> plot(pcaModel, typeVc = "x-score", parLabVc = paste0("S", 1:nrow(sacurine[["dataMatrix"]])), parAsColFcVn = as.factor(rep_len(paste0("G", 1:5), length.out = nrow(sacurine[["dataMatrix"]]))))

ADD REPLY • link written 5 months ago by etienne.thevenot • 0

It works. Thank you so much

ADD REPLY • link written 5 months ago by yuan.guo001 • 0

Similar posts • Search »