DIA-NN - automated handling of challenging data

[michaelsteidel86]

Description of feature

Dear all,

we're currently considering making use of the msquant pipeline for processing of large sample cohort DIA data.

Testing the DIA-NN GUI I took note that challenging / empty samples could be a big challenge to DIA-NN as actual processing times / file during the 1st pass (based on large predicted library) will increase several fold & run for hours eventually resulting in 0 IDs. This might be a big challenge in studies comprising thousands of files with varying data quality.

Are you aware of this problem, and are there actually automated mechanisms in place to cope with it? I.e. skipping processing if an instance runs longer than expected?

Thanks
Michael

[jpfeuffer]

Hi Michael,

you can easily set time limits for processes with a user-provided config file, however downstream processes most likely will fail. No one ever reported such a problem (ping @ypriverol), therefore we unfortunately do not know what errors/error codes to look for and also I am not sure how to tell DIANN that a certain file failed.
Maybe we can simulate an empty table but we would need feedback either from you or from the DIANN developers.

[ypriverol]

@michaelsteidel86 do you still have this issue?