sPLS function with ncomp = 1

Support
1

Hello. I want to perform an integration of two time series datasets (transcriptome and metabolome) with the timeOmics package and PLS / sPLS function.
I calculated the number of components with the getNcomp() function. And the result was 1. This worked well for the pls function.

Then I used the tuneCluster.spls() function to get the keepx and keepy number for spls(). Here I get this error message:

“Fehler in apply(X = X, FUN = function(x) { :
dim(X) must have a positive length”

If I choose ncomp = 2 the function works. There must be a problem with the number of components.
How can I fix this problem?

Thank you very much!

2

thank you @si.Arzt,
I’ll contact the maintainer of the timeOmics package and will get back to you.

Kim-Anh

3

Hi @si.Arzt,
Thank you for using the timeOmics package and pointing out this bug.
It should be fixed now.
Changes should appear within a few days on Bioconductor’s servers but you can download the patch right now using the following commands:

## install devtools if not installed
if (!requireNamespace("devtools", quietly = TRUE))
    install.packages("devtools")
## install timeOmics
devtools::install_github("abodein/timeOmics")

Antoine

4

Hi Antoine,
Thank you very much for your fast response and for fixing the bug.

I will test the package this week and get back to you if necessary.

Simon