gessulat
commented
5 years ago Hi @MarcIsak,
no worries, we are very aware of the points that you mentioned. The reason we currently limit the rescoring to one RAW file is that we need to ensure that our resources are used fairly by everyone.
Please understand that we are providing the Prosit online service free of charge. MaxQuant searches including multiple RAW files take a substantial amount of space as well as computing time on our servers. We need to temporarily store results so that users can download it. Managing this for tasks that include say 100 x 1GB RAW files is challenging and costly. Also, such a large task would block our queue for a long time, preventing other users from using our service. That is what we want to avoid.
What you can do if you have multiple raw files is starting a new task for each of it to re-score each raw file separately.
MarcIsak
vesgadj72
tkschmidt
Hi,
as I understand, when using the rescoring functionality of PROSIT, the submitted msms.txt file must come from a MaxQuant search where only a single RAW-file's been run. I think it is a pity though, because it is not very often it is useful to run a MaxQuant search with only one RAW-file. I hope I don´t sound too negative here, I just want to help out.
The rescoring function would be more useful if it could: (1) Improve a standard database search where several RAW-files been searched in a quantitative workflow (like LFQ). (2) Create a deeper and more accurate DDA-based spectral library. In this case, several RAW-files would be set up as fractions of a sample in a MaxQuant search, resulting in a single msms.txt file that could go into the PROSIT rescoring. This is what I would like to use the rescoring for, if possible.
I understand that problems can arise if one tries to rescore a MaxQuant search where RAW-files were acquired with different instrument methods or instruments. That could screw up the CE-calibration. But if all files were acquired on the same instrument, with the same instrument method I don't see why it would be a problem to include several RAW-files in the msms.txt. In many (if not most cases) the RAW-files used in a MaxQuant search are acquired by the same instrument method and instrument. So perhaps, one could just warn users of the rescoring function to not upload a msms.txt file where the RAW-files are not acquired by the same method and/or instrument.
Once again, I hope I didn't sound too pessimistic here. I think the idea of rescoring a MaxQuant search or any other search engine with PROSIT is superb.