Unknown Loss and Tracing Fragment Ions

laurenfields commented 3 weeks ago

Hello,

First, thank you for this fantastic resource!

I do have a question- I am working with a label I have developed for derivatizing arginine, and used it as a variable modification to build a spectral library in MSFragger. The spectral library I built has no neutral loss fragment ions (I didn't include them). I am now using the library to search DIA spectra in DIA-NN, and I have located >1000 peptides, which is great, but every single fragment ion has an unknown neutral loss. I would like to be able to reference the spectrum so that I can get an understanding of my label quality, but it is quite difficult with the unknown neutral loss. Do you know why these are showing up as unknown? Is there a way around this to locate the m/z of the ion?

If it helps, this was my input command: diann.exe --f D:\Manuscripts\2024_Citrullination_Quantiation\480_20240821\DIA\GQ1.mzML --lib D:\Manuscripts\2024_Citrullination_Quantiation\new_480_lib\libraries\exploris_lumos_formatted_lib_no_doubles_labelled_only.txt --threads 4 --verbose 1 --out D:\Manuscripts\2024_Citrullination_Quantiation\DIANN_19_out\test_v09_lib_vis\report.tsv --qvalue 0.01 --matrices --xic --var-mods 5 --peptidoforms --reanalyse --relaxed-prot-inf --smart-profiling --report-lib-info

And then for example this is what I get in the Fragment.Info column, with no non-unknown-loss fragment ions identified y4-unknown^1/532.272583;b11-unknown^3/541.6436157;y3-unknown^1/417.2456055;y5-unknown^1/619.3045654;b10-unknown^3/517.9645386;b11-unknown^2/811.961731;b3-unknown^1/308.1717224;b8-unknown^2/651.3717041;b9-unknown^3/480.2698669;b12-unknown^3/573.9945068;b13-unknown^3/628.348938;b15-unknown^3/695.7019653;

Thanks so much!

vdemichev commented 3 weeks ago

Hi,

but every single fragment ion has an unknown neutral loss

That's because FragPipe lib does not annotate loss - does not have the respective column. So DIA-NN then labels the respective fragments as 'with unknown loss'. This is fine.

Is there a way around this to locate the m/z of the ion?

But it's present in the Fragment.Info column, as you have shown?

Best, Vadim

laurenfields commented 3 weeks ago

Hi Vadim,

Thank you for your explanation. This makes sense. My confusion with the m/z comes from the report .tsv. I see that it tells me the MS2.Scan and the Fragment.Info. However it seems that it is pointing me to fragments that don't exist within that scan within the raw file. With that, I assumed maybe it is indicating fragment ions that occur through the specified retention time, however the MS2.Scan and the RT.Start and RT.Stop values happen in two distinct time frames, so I didn't know how these correspond. In my raw file, the corresponding MS2.Scan is at RT: 84.99 and the report has the RT.Start and RT.Stop from 85.419-85.923. Could you help me figure out where to locate these ions?

Thank you! Lauren

File.Name | Run | Stripped.Sequence | RT | RT.Start | RT.Stop | iRT | Predicted.RT | Predicted.iRT | Ms1.Area | MS2.Scan | Precursor.Mz | Fragment.Info -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- D:\\DIA\GQ1.mzML | GQ1 | AATVGSLAGQPLQERAQAWGER | 85.70026 | 85.41868 | 85.92334 | 81.06551 | 86.88835 | 85.57716 | 393775.5 | 92332 | 909.8049 | b3-unknown^1/244.1291809;y15-unknown^2/1064.539795;y12-unknown^2/936.4812012;y14-unknown^2/1029.02124;y18-unknown^2/1193.108521;y17-unknown^2/1164.597778;y16-unknown^2/1121.081787;y3-unknown^1/361.1830139;y5-unknown^1/618.2994385;y4-unknown^1/547.2623291;y7-unknown^1/817.3951416;y6-unknown^1/746.3580322;

vdemichev commented 3 weeks ago

Hi Lauren,

However it seems that it is pointing me to fragments that don't exist within that scan within the raw file.

MS2.Scan in DIA-NN means the index of the MS2 scan among all MS2 scans, not absolute index in the raw file. You can however easily find the right raw scan by matching the RT.

The fragment masses in the main report are the library values.

Best, Vadim

vdemichev / DiaNN

Unknown Loss and Tracing Fragment Ions #1232