Identification of unknown metabolites

January 21, 2013, 03:53:37 PM

Hello,

I am just starting to use metabolomics for my research and using xcms online first time. I am doing global metabolomics to see how metabolites change between different bacterial samples.

I have question about the identification of the metabolites, the xcms results gives me a potential list of metabolites for each hit along with its tolerance in ppm. Now because this is global profiling I am not sure what all metabolites to expect in my samples. My question is does anyone know a more specific way to filter out the metabolites and get the closest match for my peak or selecting the metabolite is only subjective.

Thanks

Re: Identification of unknown metabolites

Reply #1 – January 22, 2013, 12:05:44 PM

For untargeted metabolite profiling experiments you normally want to identify the metabolites that are dysregulated, i.e. significant p-values and fold change. These should be the most interesting/relevant for your experiment.
After filtering and selecting candidates (examining the putative hits based on exact mass; filtering by intensity, excluding isotopic peaks, etc.) you usually want to acquire MS/MS data for these metabolites to confirm ID.