I am curious about how to find proper compounds for Quality Control. When you do metabolomics research, usually you have many samples. There are different kinds of samples such as controls and samples from different treatments. Also, you usually have to do various repeats. So you can expect plenty of samples waiting to be analyzed using certain equipments, say, LC-MS. But how can I make sure the repeatability of the analysis, say, the retention time or the mass accuracy, is acceptable? Of course QC samples can help a lot in this process. So now comes the question, how to choose proper QC samples during metabolomics research? Are there any general guidelines? How to choose different QC samples for different analysis?
You can use a pool of all the samples as a QC injection. Add up all the samples in equal amounts and create a pool sample, this way you will not miss the compounds unique to your treatment as compared to control.
Another QC sample which you should run every 20 samples or so is a mix of pure standards in equal ratio, this way you can check if there is any RT shift.
Third way is to add isotopically labelled internal standard. But for untargeted analysis, the output file will be full of peaks and labelled compounds wont match or mismatch to your unlabelled library compounds.
Kind regards,
Charan
You can use a pool of all the samples as a QC injection. Add up all the samples in equal amounts and create a pool sample, this way you will not miss the compounds unique to your treatment as compared to control.
Another QC sample which you should run every 20 samples or so is a mix of pure standards in equal ratio, this way you can check if there is any RT shift.
Third way is to add isotopically labelled internal standard. But for untargeted analysis, the output file will be full of peaks and labelled compounds wont match or mismatch to your unlabelled library compounds.
Kind regards,
Charan
I suggest you reading following article: "Dunn W.B., Wilson I.D.The importance of experimental design and QC samples in large-scale and MS-driven untargeted metabolomic studies of humans. Bioanalysis 4 (2012) is 18 p 2249"
Depending on your goals (untargeted or targeted) I would recommend pooled QC-samples (collecting aliquots of every sample to be analysed) or reference standards (where you know the m/z of each molecule) and monitor/correct for any shift in retention time or mass accuracy.
All of the previous answers. But also: (a) cross-check both QC and real samples with NMR (non-binning), it has far superior quantitative performance compared with mass spec, and this well-known fact (there are papers showing superiority to even isotope-ratio MS) is much ignored; (b) try to arrange some real sample runs under extremely-high resolution MS (>200,000 resolving power @ 400 m/z) to look for m/z interferences, as even the best QC is worthless if the MS and MS/MS is mis-identifying the compounds & fragments (c) "metabolite" is an INTERPRETATION, NOT A RESULT OR AN ANALYTE. For example, your analysis of lactate can be a contamination not from your experiment, thus you canNOT analyze for "metabolites", only compounds. This leads to the fourth and most important "QC" item is PROPER BLANKS & CONTROLS. This is not just analytical blanks & controls, but those that ran the gamut of the experiment. If it is cell culture, you need blank plates with no cells. If it is banked specimens you are analyzing, you must make sure there are blanks (such as empty blood tubes) - but keep in mind that in almost all cases, such critical QA/QC samples simply don't exist, so if you have an input, make sure that this designed into the study.
- Badges
- Science topic
- Similar topics
- Chemistry
- General Biochemistry
- Proteomics