Skip to main content
Topic: Possible bug when groupCorr called with xraw argument (Read 6447 times) previous topic - next topic

Possible bug when groupCorr called with xraw argument

G'day,

When calling groupCorr with the xraw argument set I see the following error message:

Code: [Select]
> groupCorr(object=yanGI, calcIso=TRUE, xraw=xcmsRaw(file))
Start grouping after correlation.
Error in peaks(xa@xcmsSet) :
  error in evaluating the argument 'object' in selecting a method for function 'peaks': Error: object 'xa' not found

Looking at the source code in fct_groupCorr.R it appears the xa has not been assigned a value.

Code: [Select]
  # Check LC information and calcCorr was selected
  if(calcCiS && object@xcmsSet@peaks[1,"rt"] != -1){
    if(!is.null(xraw)) {
      #Use provided xcmsRaw
      maxscans <- length(xraw@scantime)
      scantimes<-list();
      scantimes[[1]] <- xraw@scantime
      pdata <- peaks(xa@xcmsSet)
      EIC <- CAMERA:::getEICs(xraw,pdata,maxscans)

Is this a bug?

Regards,
Chris.

Re: Possible bug when groupCorr called with xraw argument

Reply #1
Hi Chris,

yes, this is indeed a bug.
I fixed it in 1.11.10, which should be available on the BioConductor server tomorrow.

Thanks for pointing this out.

Carsten

Re: Possible bug when groupCorr called with xraw argument

Reply #2
G'day,

Quote
I fixed it in 1.11.10, which should be available on the BioConductor server tomorrow.

It doesn't seem to have appeared (at least update.packages("CAMERA") does nothing, and the CAMERA page says 1.10.0).

Thanks,
Chris.

Re: Possible bug when groupCorr called with xraw argument

Reply #3
BioC automatically builds packages over night so give it a little more time to show up.
If you need it immediately you can check out a copy through SVN.