Metabolomics Society Forum

Software => R => XCMS => Topic started by: kbladon on February 16, 2015, 04:22:05 AM

Title: Peak Detection Errors
Post by: kbladon on February 16, 2015, 04:22:05 AM
Hi all,

I am using Bentobox (v3), and get the error message below when running the peak detection script using centWave.
'Error in checkForRemoteErrors(val) :
  89 nodes produced errors; first error: m/z sort assumption violated ! (scan 618, p 7536, current 951.7201 (I=35.00), last 951.7378)'
traceback()
4: stop(count, " nodes produced errors; first error: ", firstmsg)
3: checkForRemoteErrors(val)
2: xcmsClusterApply(cl = snowclust, x = argList, fun = findPeaksPar,
      msgfun = msgfun.featureDetection)
1: xcmsSet(method = "centWave", peakwidth = c(CPWmin, CPWmax), ppm = ppm,
      snthresh = xsnthresh, lock = LM, integrate = integrate, nSlaves = nSlaves,
      prefilter = c(5, 1000), noise = (300), scanrange = c(30,1363))

I have seen that many people have had this problem due to conversion software.  This data was originally converted using DataBridge (Data was acquired on Waters Synapt G2).  I reconverted using databridge on another computer is case it was a system issue, with no success.

As suggested in other feeds, I then converted the raw files to the mzXML format using Proteowizard, and get the following error message:
 '89 nodes produced errors; first error: Time for scan 1499 (1319.56) greater than scan 1500 (1.225)'
> traceback()
4: stop(count, " nodes produced errors; first error: ", firstmsg)
3: checkForRemoteErrors(val)
2: xcmsClusterApply(cl = snowclust, x = argList, fun = findPeaksPar,
      msgfun = msgfun.featureDetection)
1: xcmsSet(method = "centWave", peakwidth = c(CPWmin, CPWmax), ppm = ppm,
      snthresh = xsnthresh, lock = LM, integrate = integrate, nSlaves = nSlaves,
      prefilter = c(5, 1000), noise = (300), scanrange = c(30,
          50))
My run time is 22 mins, so (from my naive eye) it looks like its either replacing the last scan with the first, or looping.  I have tried playing with the scan range used (30-1363 is default), but it all seems redundant as I get the same error every time.

Does anyone know a successful way of fixing this issue?  I know this issue has been posted a few times before, but none of the current suggestions have worked when applied.

Thanks in advance,
Kallie
Title: Re: Peak Detection Errors
Post by: Jan Stanstrup on February 17, 2015, 02:42:36 AM
Tried this: viewtopic.php?t=144&p=274 (http://www.metabolomics-forum.com/viewtopic.php?t=144&p=274) ?
Title: Re: Peak Detection Errors
Post by: kbladon on February 17, 2015, 04:54:12 AM
Hi Jan,

Yes, I've tried this with no success

Kallie
Title: Re: Peak Detection Errors
Post by: Jan Stanstrup on February 17, 2015, 06:19:40 AM
Perhaps try masswolf then: viewtopic.php?f=26&t=359&p=1694 (http://www.metabolomics-forum.com/viewtopic.php?f=26&t=359&p=1694)