How to increase processing speed for phosphoproteomics data in DIA-NN?

[demhibou]

Dear Vadim,

First, thanks for making a great software! We are generating DIA phosphoproteomics data using a Bruker timsTOF HT on human cells infected with viruses. We have generated an in-silico spectral library in DIA-NN with the attached settings (see attached library generation settings image) and used this to search our .d files (see attached settings used for the search image, which were slightly refined). Our .d files are often around 3GB each.

The spectral library generation took about 12 hours, which is OK for us as we only have to do this once per project. However, searching our data is taking >12 hours per sample, even when we load in this pre-created spectral library.

We are wondering if you have any advice for how we can refine our settings to greatly enhance the speed of our phosphoproteomics searches.

Thank you!

[vdemichev]

Hi,

We have generated an in-silico spectral library in DIA-NN with the attached settings

Please change precursor charge range to 2-3 as recommended here https://github.com/vdemichev/DiaNN?tab=readme-ov-file#ptms-and-peptidoforms

However, searching our data is taking >12 hours per sample

Can you please share the logs for both the library generation and the analysis?
Please note that it's OK to use the ultra-fast mode for phospho on timsTOFs.

Best,
Vadim