I am using xcms for metabolomic work, and I'm coming across a bug that I can not seem to track down. I've scoured the online forums and am trying a grid search of various parameters, but I'd love to get your input on this problem.
To identify features in an experiment with various samples that are independently injected into the mass spec and correct for retention time.
The commands I am using (in the order that I am using them):
Reading in the data: readMSData
Identify chromatographic peaks: findChromPeaks() with a CentWaveParam object
Retention time correction: adjustRtime() with a ObiWarpParam object
Grouping peaks: groupChromPeaks() with a PeakDensityParam object
Output: featureValues() to list all features with each column representing the intensity of the features in one sample.
All the features identified in the one sample have the very similar intensities (+/- 10%) in the featureValues() output. said differently: Each feature has different intensities in all samples, but all features in the same sample differ by approximately 10%.
This trend is not consistent with the raw traces of the data. I see much more variability between peak intensities in each sample.
I've been reading the xcms documentation to track down what could be changing/controlling peak intensities, but I can not seem to find any parameter in the commands I'm using that could affect peak intensities. I'm very much working in the dark here. Any insight on your part would be HIGHLY appreciated.