search

OpenMS / OpenMS

The codebase of the OpenMS project
https://www.openms.de
Other
478 stars 318 forks source link

Non-fatal errors in comet #7644

Open timosachsenberg opened 1 day ago

timosachsenberg commented 1 day ago

Reported by user: I’m encountering a few non-fatal errors related to modifications despite trying different configurations in the Comet parameters file and the CometAdapter command. These errors have been on my mind, and I wonder if the high number of unmatched hits might be related to this issue. I’d really appreciate any guidance. Here are some of the errors I’m seeing: Non-fatal error while loading '/tmp/20241101_180330_lambda01_1301970_1/result.pep.xml': Errors during parsing of aminoacid/terminal modification element: Non-fatal error while loading '/tmp/20241101_180330_lambda01_1301970_1/result.pep.xml': No modification description given. Trying to define by modification mass. <Non-fatal error while loading '/tmp/20241101_180330_lambda01_1301970_1/result.pep.xml': Errors during parsing of aminoacid/terminal modification element:> occurred 3 times Non-fatal error while loading '/tmp/20241101_180330_lambda01_1301970_1/result.pep.xml': Modification '243.029629' is not uniquely defined by the given data. Using 'O4'-(phospho-5'-RNA)-L-tyrosine (Y)' to represent any of 'O4'-(phospho-5'-RNA)-L-tyrosine (Y), O4'-(phospho-5'-RNA)-L-tyrosine (Y), O4'-(phospho-5'-DNA)-L-tyrosine (Y), O4'-(phospho-3'-DNA)-L-tyrosine (Y), Phospho (Y)'. Thank you very much for your help.

jpfeuffer commented 1 day ago

We would need the pep.xml to check. The problem is probably that they use the params.xml and not the settings of our adapter. We have no way of uniquely associating the modification masses to modifications in that case.

As a side note: O4'-(phospho-5'-RNA)-L-tyrosine (Y), O4'-(phospho-5'-RNA)-L-tyrosine (Y) appearing twice looks very fishy. Considering that the DNA mod has 5' and 3' I am wondering if that is an error in our DB.

jpfeuffer commented 1 day ago

Also, maybe we should prioritize protein modifications in lookup situations? Feels like this might have worked before but after introduction of RNA mods this will not be compatible with special Phospho scoring tools that look for the normal "Phospho" mod.

timosachsenberg commented 1 day ago

how does this relate to: https://github.com/OpenMS/OpenMS/issues/5109 ?

jpfeuffer commented 1 day ago

My fix that addressed this issue only works if you specify the mods on the command line. Not via a params.txt (which this user most likely did)

SR-46 commented 1 day ago

Thank you all for your guidance. Based on @jpfeuffer's comment, I removed all modifications (both fixed and variable) from the parameter file, and this resolved the errors I was encountering.

The CometAdapter webpage mentions: “Comet settings not exposed by this adapter can be directly adjusted using a param file, which can be generated using comet -p. By default, all (!) parameters available explicitly via this param file will take precedence over the wrapper parameters.”

So, initially, I only used the parameter file for modifications and other settings, but this led to errors. I then tried specifying modifications in both the param file and the CometAdapter command, which resulted in even more errors. However, removing all modifications from the parameter file eliminated the errors although significantly increased the number of unmatched peptides, which I think is reasonable.

Thank you very much for your help.