Empty files produced by kevlar split cause crash of kevlar assemble

[serosko]

Empty files produced by kevlar split cause crash of kevlar assemble

[serosko]

Yes, I found the option after writing the issue and tried it with "numsplit": 4.
Now that I have the results, I can confirm that this fixed the issue.

[standage]

Thanks for the report. Did you use the default "numsplit": 16 in your config file? As you've probably guessed, this is used to split the data and compute some later steps of the workflow in parallel. For 30x human WGS data, this will provide a decent speedup. For a mock 25 Mb genome with 10 de novo variants I don't expect it to make any practical difference. Try running it with "numsplit": 2 (or even 1) and see if that fixes the problem. If so, I'll add a clarifying note to the tutorial.

[standage]

Roger that. I'll close this thread once the documentation has been clarified.