DBsplit options for large genomes

[AdamStuckert]

We are assembling a genome that is > 6 Gbp in size, and we are wondering what the best DBsplit flags are for our purposes. The readme says:

pa_DBsplit_option=-x500 -s200
ovlp_DBsplit_option=-x500 -s200

# small genomes (<10Mb)
pa_DBsplit_option = -x500 -s50
ovlp_DBsplit_option = -x500 -s50

Does this mean that everything > 10 Mb should have the x500 -s200 option, or should this be adjusted for much larger genomes?

Thanks in advance!

[pb-cdunn]

-x500 -s200 is fine for even larger genomes. But be aware that the memory requirement for daligner will be 16x higher with -s50 than with -s200, as it does block-to-block comparisons.

[harish0201]

Hi!

I'm sorry to necro-bump this issue up, but I was wondering about the same, is there a way to prune the folders/files created during Falcon?

So, far the daligner-runs have finished and I'm looking at nearly 13Tb worth of files in a 16Tb drive.

pa_DBsplit_option=-x500 -s400
ovlp_DBsplit_option=-s400

Is there are parameter otherwise that will prune the files generated?

[pb-cdunn]

Try

falconc pbcromwell-rm-las -h

That is designed for use with our Cromwell workflow, but it should work for you too. Be sure to run in "dry-run" mode first.