Quality of MS/MS data?

[rick1112]

Hi
Many of you work on mass spectrometry dealing with multitude of application.
From your experience how you all go about assessing Quality of MS/MS spectra generated? What criteria is used to assess if MS/MS spectra is acceptable, and result obtained are provides confidence?

[Rndirk]

That's a broad question. It depends on the analysis. Do you have the freedom to set your own limits, or is the analysis regulated? In the second case you stick to the rules.

MS/MS assessment of signals is not so different from MS. Retention time, sensitivity (detection limit, signal-to-noise,...) are interpreted in the same way. We usually measure 2 transitions for each analyte, 1 is used to quantify and the other is the 'qualifier'. The ratio of qualifier to quantifier is compared between signals in samples and in standards. This ratio can be 20% different in samples compared to standards, for example.

[rick1112]

Hi Rndirk
Thanks for the replay
I don’t want to set any limits on my data rather I won’t like to know what all aspects are looked into to assess the quality of MS/MS data (BTW I work on proteins and peptides ). From my understanding quality of MS/MS data of a peptide has a huge impact on the correct identification of the peptide and its modifications.
Was wondering how you all look into MS/MS data ?

[lmh]

My view is that MS/MS spectra are like any other form of data. The key thing is whether they can be repeated. I dislike trusting single MS/MS spectra from data-dependent methods that attempt to cover as many precursor ions as possible.
If an ion shows up in multiple repeat MS/MS spectra across the chromatographic peak, it is almost certainly a genuine fragment and not just noise. If it shows up in just one, it may be a fragment, but it's higher-risk. If the spectrum looks different every time it's collected, and is weak, the whole thing is probably noise and a load of rubbish.