[mira_talk] Re: exiting mira without error message
- From: Chenling Antelope <chenlingantelope@xxxxxxxxx>
- To: mira_talk <mira_talk@xxxxxxxxxxxxx>
- Date: Wed, 14 May 2014 17:55:50 -0700
Thanks Bastien for the answer!
However I am currently unable to correct my reads because I lack the glib
version required by celera.
Also, I used miramem to estimate the RAM required, which is a lot smaller
than my actual RAM 16G
************************* Estimates *************************
The contigs will have an average coverage of ~ 35.0 (+/- 10%)
Estimates may be way off for pathological cases.
RAM estimates:
reads+contigs (unavoidable): 1.1 GiB
large tables (tunable): 2.8 GiB
---------
total (peak): 3.9 GiB
add if using -CL:pvlc=yes : 432. MiB
Note that some algorithms might try to grab more memory if
the need arises and the system has enough RAM. The options
for automatic memory management control this:
-AS:amm, -AS:kpmf, -AS:mps
Further switches that might reduce RAM (at cost of run time
or accuracy):
-SK:mhim, -SK:mchr (both runtime); -SK:mhpr (accuracy)
So I added in option -GE:amm=on -GE:kpmf=10 and it still stopped at the
exactly the same point of the analysis.
Is there something else I can do to trouble shoot?
Best,
Chenling
2014-05-14 15:13 GMT-07:00 Bastien Chevreux <bach@xxxxxxxxxxxx>:
> On 14 May 2014, at 23:35 , Chenling Antelope <chenlingantelope@xxxxxxxxx>
> wrote:
>
> I am trying to assemble a genome with pacbio reads. My reads are
> uncorrected. and I have in total 28245 reads.
>
>
> Running MIRA with uncorrected reads is not recommended.
>
> Mira ran for about half an hour before it crashed.
>
>
> Your output lets me think that it wasn’t MIRA which crashed, but rather
> the OS seems to have killed it. I suspect because of memory problems, i.e.,
> not enough RAM.
>
> B.
>
>
>
Other related posts:
