Semi-specific digestion type

[She1111]

Hi Vadim,

I am doing limited-proteolysis in proteome. As I use MaxQuant to analyse DDA data, I select semi-specific as digestion type and it will run well with this option. Because I find low amount of semi-specific peptides in report data from DIA-NN, I wonder whether DIA-NN has this function when it generates spectral library in silico? I use report.tsv to analyse. Dose stripped_sequence and pg_max_lfq refer to peptide sequence and intensity (like sequence and intensity in MaxQuant) respectively? Thank you so much for your great effort!

Best regards, Shel

[She1111]

Hi,

237 has great questions and answers! Maybe I could try Fragpipe at first!

Best regards, Shel

[vdemichev]

Hi Shel,

With a DDA-based library, DIA-NN does not really need extra settings, will just search all peptides in the library fully automatically. In lib-free mode, I would suggest not to use semi-specific digest, as this would inflate search space a lot.

I am not sure, why would you expect the enzyme to cleave non-specifically in limited proteolysis?

Best, Vadim

[She1111]

Hi Vadim,

Thank you for your reply! I would like to expect semi-specific digestion as I used proteinase K (non-specific enzyme) for limited proteolysis then typsin/LysC for completely digestion. It seems using DDA-based spectral library is more feasible in this case :).

Best, Shel

[She1111]

Hi @vdemichev,

I used fragpipe to generate spectral library (of course in semi-specific digestion mode) for DIA-NN and it worked well! But the identifications of both protein and peptide are much lower than the results which use in-silico spectral library for searching. I noticed you suggested to add desirable peptides into FASTA to build a in-silico library in a previous issue. So I added semi-tryptic peptides produced from DDA data set to FASTA and process the further steps. Identifications are higher, semi-tryptic peptides could also be detected. But I am wondering whether it could have risks or issues to do this?

Best, Shel