[mira_talk]
- From: Torben Nielsen <torben@xxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Fri, 20 Dec 2013 14:16:25 -1000
I’m learning - slowly - to stick with MIRA defaults. I’ve been “regressing” for
some time now. First I tried to read the documentation several times and I
thought I understood how to intelligently set the parameters. Wrong. So one by
one, I took my special settings out. And now I’m down to one last bastion (not
the same as “bastien” mind you) of resistance. Well, two actually….. mrs and
pr. On my first read, I confused those with the quality of the assembly….
The documentation clearly says “I propose a maximum of 80 here, but I disposed
and made it 95….. And the results were still good enough that I was more
inclined to celebrate than weep. But then I tried 80 and life was ever so much
better. The question is how much better can it get? Bastien says that things
will blow up at 55 or below. Does that mean I can go to 60 and still be good?
I have 384 GB or memory; i.e., it’s not a small machine. I’m also willing to
let it run for several weeks.
Suggestions? Anyone has experience in this? This is metagenomes of the worst
kind. It’s time series of metagenomes spanning several years. And the samples
are taken where the Sun doesn’t shine and even NCBI knows next to nothing.
Mesopelagic.
Thanks, Torben
--
You have received this mail because you are subscribed to the mira_talk mailing
list. For information on how to subscribe or unsubscribe, please visit
http://www.chevreux.org/mira_mailinglists.html
Other related posts:
