You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It is possible to run Sarek repeatedly for the same sample or different samples using the same output-folder (results-folder), and some users do run Sarek like that.
Currently there are a number of issues with reusing the results-folder.
The multiqc-data gets overwritten,
joint-germline-data gets overwritten,
csv/variantcalled.csv gets overwritten, (there is probably more files that get overwritten)
if starting more pipelines at the exact same time, then different pipelines may construct pipeline-info-files with exactly the same name. (Tower actually did this at DNGC; bug reported to Rob Syme and Harshil Patel.)
Using the same results-folder for different pipeline-runs makes it very difficult to subsequently, say, find and delete all files from some particular pipeline-run.
Should we try to solve those issue?
Perhaps we just want to advise against using the same results-folder for different pipeline-runs?
N.B. What about work-folders? On re-runs (-resume) one, of course, wants to use the same work-folder, but what about work-folders for pipeline-runs of different samples? I see no reason that they should share work-folders. (I know that the subfolders with the hash-strings make any clashes/conflicts between pipelines sharing the same work-folder very unlikely, but still?)
The text was updated successfully, but these errors were encountered:
Description of feature
It is possible to run Sarek repeatedly for the same sample or different samples using the same output-folder (results-folder), and some users do run Sarek like that.
Currently there are a number of issues with reusing the results-folder.
csv/variantcalled.csv
gets overwritten, (there is probably more files that get overwritten)Should we try to solve those issue?
Perhaps we just want to advise against using the same results-folder for different pipeline-runs?
N.B. What about work-folders? On re-runs (
-resume
) one, of course, wants to use the same work-folder, but what about work-folders for pipeline-runs of different samples? I see no reason that they should share work-folders. (I know that the subfolders with the hash-strings make any clashes/conflicts between pipelines sharing the same work-folder very unlikely, but still?)The text was updated successfully, but these errors were encountered: