Problem with LOESS batch correction metanode

I am having problems using the LOESS metanode on Knime. When I open up the node, the error seems to occur on the R view workspace.
I am getting the message:

ERROR R view (Workspace) 0:828:0:584 Execute failed: Error in R code:
Error: unused arguments (printL= FALSE, plotL = FALSE)

The line of the script that seems to cause this error is:

plotBatchF(rawMN, samDF, spnN)

I have attached some screenshots here.

Any help on how to resolve this error would be greatly appreciated!

Thank you


Could you provide us with a link to this metanode or maybe even a sample workflow that would show the problem without spelling any secrets.

Hi ,

Yes please see attached workflow
Drinks_LOESS.knwf (2.9 MB)

Thank you

@am221 this is an absolute mess to debug. But based on what the error message says what is possible is that deep in the code deprecated functions from a (?) Bioconductor package are being used that might have to be fixed.

pcaLs <- opls(pcaMN, predI = 4, algoC = "svd", printL = FALSE, plotL = FALSE)

The last two options seem to have been deprecated from the package and might no longer be supported - and possibly create the error message.

image

I have no idea what will happen if you just delete them or you might have to replace them with new settings depending on what you want to do.

As an additional remark: the workflow in its current state cannot be run by anyone without the data and even then there are several loose ends and so on. So there is absolutely no guarantee that this point will fix the whole things since the workflow has lots of R scripts and dependencies so something else might be lurking. If you are the creator of this workflow and know what you are doing: fine. Otherwise proceed with extra caution.

1 Like

Dear mlauber71,

Thank you for your help. Your advice seems to have done the trick regarding the unused arguments error message but now I am getting this message:

Any ideas?

BW

@am221 non other than what the error message says. You will have to check the number of rows and items that go into the function.

Maybe you extract the data from the R node before and check the values. But as I said. The whole workflow seems complicated and you should make sure you understand what these functions and packages do.

1 Like