xcms
xcms copied to clipboard
fillPeaks() replaces NA values with large values
Hi,
I am trying to use fillPeaks() to fill in NA values in my dataset but I am seeing very large filled in values. For example, the filled values are around 20,000 - 40,000 when a lot of the real peaks have values of 500-20,000. The original mzXML files were first processed using xcmsSet(), retcor(), and group().
Any ideas why this could be happening? Any feedback would be appreciated!
Thanks in advance. Arjana Begzati
Can you check an EIC of the filled peak ? Which values are you using in the groupval, e.g. into ? Yours, Steffen
In addition, I would strongly suggest that you switch to the new functionality (see the xcms vignette for details), as we did also some improvement/changes especially in the new code for the filling-in.
I just checked with the developers of geoRge, the package I am ultimately trying to use, and it does not support xcms new output object.
Could you guide me on how I can view the chromatogram of a specific value in the groupval(grouped_rtcor_xset, value = "into") output? I am trying to check the filled in values that seem pretty high.
Thanks!
I just saw on https://bioconductor.org/packages/devel/bioc/vignettes/xcms/inst/doc/xcms.html#1_introduction that is should be possible to convert a XCMSnExp into a xcmsSet object using xset <- as(x, "xcmsSet"). This doesn't work for me; I get "Error in as(raw_data, "xcmsSet") : no method or default for coercing “OnDiskMSnExp” to “xcmsSet”". This same error occurs when I try to do the conversion for my data set and the example from the above mentioned tutorial.
It would be great if I could do this conversions since for my downstream analysis (geoRge & x13cms) it seems like the xcmsSet object is the only possible input.
Thanks in advance!
Firstly, if you are using xcmsSet, retcor and groupval you are using the old xcms user interface and classes. In that case you don't have to use the as(x, "xcmsSet") because you have already an xcmsSet object.
If you used the new objects, i.e. you read the data first with readMSData, then you do findChromPeaks etc you'll have the new XCMSnExp object. That one you would have to convert to an xcmsSet with the as(x, "xcmsSet").