We are new in using XCMS and we have some questions. -Is there a package in R that can be used for peak deconvolution?
-Once you have a XCMS peak table or a table made with CAMERA, how do you indentify the different metabolites? Is their an automated tool, where you can upload your table, which then identifies the different compounds and puts a name on it?
Our lab determines metabolite profiles in human feces using GC-MS (untargeted). Until now, we analysed every chromatogram ourselves, determined which metabolites were present in each sample and then made a peak intensity table ourselves with all metabolites found in the samples belonging to one study. We want to do this now in a more 'automated' way and found that a lot of people are using the XCMS package in R. As we are not confident with the XCMS package, we encountered some problems, when the peak intensity table is generated.
In the tables, that we generate ourselves, we only assign one mass (the mother ion) with a certain retention time to one metabolite. In the peak intensity table generated in the XCMS package, we found several masses with the same retention time, corresponding to the same metabolite. It means that from the 300 compounds that were detected in our samples, there will be only 100 left when going through the table and putting together all features belonging to the same metabolites. Is their a kind of filtering or control that we can apply on the table so that in the end we only have one mass (the most important one) and the retention time belonging to one metabolite?
The program we are using now to look at our chromatograms is Xcalibur. With this program we are able to deconvolute our chromatograms in that way that we can define overlapping peaks . If we look into the XCMS peak intensity table, the program only detects one peak (the one with the biggest AUC), but is not able to detect other peaks lying under this bigger peak. Is there a way to look for smaller peaks that are overlapped by bigger peaks?
