[mira_talk] Re: empty results folder

On Friday 20 February 2009 Rex Malmstrom wrote:
> I am trying to get MIRA up and running, but I'm have some difficulty.
> After running a de novo assembly of 454 FLX data, I have an empty
> *_results directory.  The *_info_assembly.txt  was created and
> indicates the sequences were assembled and contigs were formed, but
> there are no files in the _results directory.  Any ideas?  Thanks.

Hi Rex,

I've taken a peek at the log file and the directory listing: MIRA stopped in 
the midst of the fourth pass and did not finish correctly. As the STDERR output 
was not redirected to the log file, I can only speculate on the reason.

Assuming that it was not a simple CTRL-C or other kill, the only plausible 
thing I can think of is that you are running the 32 bit version and that you 
encountered an out of memory error along the way. The reason why I think that 
this is the case is because you've got some really high coverage (~180x), 
which is good per se but becomes a problem as one of the tables grows to sizes 
(2.5 GiB)  that is dangerous for a 32 bit machine.

There are several possibilities to remedy the problem and tell MIRA to use 
less memory. The best one being a version with some new algorithms that take 
care not to lead to an explosion of memory usage. But I'll contact you by mail 
for that.

Regards,
  Bastien


-- 
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: