“in_progress” decoy files in some docking2 local_docking job tar archives

[Anonymous]

In one of my ROSIE docking2 local_docking jobs (17702), I noticed that its tar archive’s outputtrigger-00001.dock folder contained several files like proteins_0130.in_progress.bz2 with the unusual size of 42 bytes in addition to many files like proteins_0129.pdb.bz2 and proteins_0131.pdb.bz2 with the more typical size of 82-84 kbytes. Do you know what causes these “in_progress” files to occur and how I can prevent them from occurring? Is the cure as simple as re-opening the job’s web page and then reloading its tar archive file?

Thanks!

[Anonymous]

The in_progress files are used by Rosetta in certain multi-processor settings as an internal annotation of which processor is working on which output structure. Normally they’re cleared up when the processor outputs the final model. However, if the job is killed by the clustering system (or dies with certain errors), those files don’t get cleared up, and will make it to the end results.

Assuming  that there’s nothing about your protein system and settings that’s causing  major issues with the protocol (in which case you’d see a *lot* of these files), there’s nothing really on your end to prevent them. You can safely ignore them, though. They shouldn’t have any impact on the results of the jobs which did complete successfully.

[Author]

Posts