ions in the same pseudospectrum not identified as isotopes

[peterdaniel666]

Hi,

I've beeing using CAMERA 1.45.2 to process LC-MS QTOF data. The following codes were used to exact isotope peaks.

xs.s <- xcmsSet(file, method = "centWave", ppm = 50, peakwidth=c(5,20),noise = 20000, snthresh = 10)     
xs.s <- xsAnnotate(xs.s)  
xs.s <- groupFWHM(xs.s, perfwhm = 0.6)   
xs.s <- findIsotopes(xs.s, ppm=50, mzabs = 0.01, filter = FALSE)  
peaklist <- getPeaklist(xs.s)  

I noticed in the peaklist, there are two ions in the same pcgroup (I assume same pcgroup indicates the same pseudospectrum), mz 514.312 maxo 1.5e6 and mz 514.814 maxo 1.1e5. As I understand, these two ions in the same pcgroup with a difference of 0.502 should be identified as isotopes of a [M]2+. However, that's not the case. When I ran xs.s <- groupFWHM(xs.s, perfwhm = 2) instead of xs.s <- groupFWHM(xs.s, perfwhm = 0.6) , less pcgroups were identified, the same two ions remained in the same pcgroup, but this time, they were identified as isotopes of a [M]2+, and this [M]2+ has no other isotopes.

Such results are very confusing to me. The C12/C13 ratio filter was off and the two ions remained in the same pcgroup, why does perfwhm affect the isotope identification?

Thank you very much.