Skip to content
This repository has been archived by the owner on Aug 30, 2023. It is now read-only.

Prosit rescoring on own server #92

Open
mezarino opened this issue Nov 8, 2022 · 1 comment
Open

Prosit rescoring on own server #92

mezarino opened this issue Nov 8, 2022 · 1 comment

Comments

@mezarino
Copy link

mezarino commented Nov 8, 2022

Dear Siegfried or Tobias,

Thank you for your support! I have a question: can you provide more detail on how to perform the rescoring of predictions made by Prosit from a peptidelist.csv file (derived from a MaxQuant msms.txt file) via Percolator? Especially, given the fact that Prosit doesn't output any cosine similarity to the MS2 spectra (which I understand is obvious since the RAW file is not an input of Prosit).

I'm referring here to a previous question from another user to whom you provided a very short answer. It would be great if you can pinpoint me to some tutorial or some detailed explanation. Thanks!

Best,
Victor

Hi Marc,

you can download the code here and run it locally, but you would need a server with an NVIDIA GPU. Here we only provide the prediction tool though. For the rescoring you would need build prediction based features and run a percolator on that files to rescore them.

Originally posted by @gessulat in #16 (comment)

@polklin
Copy link

polklin commented Apr 12, 2023

Hi @mezarino,

I'd like to use Prosit for rescoring locally as you was trying to.
Have you found a way to build features from Prosit output, that you can feed to Percolator ?

Best,
Paul

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants