DIA-NN output for annotated spectra generation

[joachims1956]

Dear DIA-NN community,

Thanks for curating this Github and further DIA-NN developments.
I am looking for the easiest way of getting annotated spectra to validate the identification of single-hit proteins. Therefore, I would like to match library spectra with the experimental acquired spectra of the peptides of those single-hit proteins.
Can this be done in DIA-NN? Do I need to add addition commands in DIA-NN to create the needed output? Can this be done in DIA-NN combined with e.g. Skyline?

Thank you very much.

[vdemichev]

Hi Joachim,

Yes, you can use DIA-NN Viewer (integrated in the GUI). Also, there will be updated Skyline integration in the next DIA-NN update.

Best,
Vadim

[vdemichev]

Is there a way to get the annotated spectra directly; meaning as fragment intensities over fragment mass/charge ratio?

Yes, that's what in the XIC files (not just spectra, but also how they change during peptide elution).

Can the xic.parquet files be filtered for single-hit protein peptides?

You can cross-reference them to the main report which has information on peptide to protein matching.

Does the DIA-NN viewer shows the fragments of the experimental spectra matching the fragments of spectra of the library?

Either library ones or you can also specify theoretical ones. Please see https://github.com/vdemichev/DiaNN?tab=readme-ov-file#visualisation, I also suggest to just quickly run it on the data recommended here https://github.com/vdemichev/DiaNN?tab=readme-ov-file#getting-started-with-dia-nn and examine the .parquet files with TAD Viewer https://www.tadviewer.com/

[joachims1956]

Ok. Thanks again. So regarding the fragments of the library, I guess it shows the fragments of a given peptide in the viewer that were used for the given peptide identification and not all fragments of a given peptide in the library.

Thanks and best.

[vdemichev]

All fragments with non-zero signals

[nkdailingyun]

Hi Joachim,

Yes, you can use DIA-NN Viewer (integrated in the GUI). Also, there will be updated Skyline integration in the next DIA-NN update.

Best, Vadim

Dear author of DIA-NN,

I have a similar issue regarding the original question and hope to get help here.
I have used version 1.8.1 to process some data and need to validate the identification of single-hit proteins. I am wondering how to visualize the spectra? It seems that XIC was not yet enabled in the 1.8.1 version. I am aware that the new DIA-NN 2.0 has a nice Viewer feature of integrating with Skyline. Thanks for the improvement. However, in this case, I hope to get around with the old search result from 1.8.1.

Thanks a lot.

[vdemichev]

DIA-NN 2.0 has a nice Viewer feature of integrating with Skyline

Also has a very powerful built-in XIC viewer.

However, in this case, I hope to get around with the old search result from 1.8.1.

I think you can import the .speclib file it generated from the empirical library it created as part of MBR into Skyline. This will be the only way as 1.8.1 does not output XICs.