Metabolomics Society Forum

Software => R => XCMS => Topic started by: wedmands on October 19, 2011, 08:28:34 AM

Title: centWave sort assumption violation
Post by: wedmands on October 19, 2011, 08:28:34 AM
Hi,

I have been having trouble with the centWave peak picker I continually get the following error message at the very end of the process?
Error in checkForRemoteErrors(val) :
  100 nodes produced errors; first error: m/z sort assumption violated ! (scan 611, p 3578, current 322.1408 (I=15.85), last 322.1423)

I have tried peak picking with each individual sub folder in turn to see if one of the folders contains a corrupt file or maybe something else but I still get the same error.

Using the default XCMS not centWave it seems to work fine strangely. I also converted my files using the Databridge programme to netCDF files if this is pertinent information.

This is the command I utilised I set the CPWmin, CPWmax and snthresh beforehand.
>peakmatrix <- xcmsSet(method="centWave", peakwidth=c(CPWmin,CPWmax), ppm=ppm, snthresh=xsnthresh, nSlaves=8)

Apologies for back to back help requests I am new to R and using XCMS and I am not a computational biologist by any means. I wonder if anyone has any ideas?

Will
Title: Re: centWave sort assumption violation
Post by: Ralf on October 19, 2011, 09:51:14 AM
Please try to convert your files using proteowizard as described here (http://https://xcmsonline.scripps.edu/docs/fileformats.html)
and see if the problem persists.