[mira_talk] Re: megahubs ?
- From: Jan van Haarst <jan@xxxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Mon, 2 Mar 2009 07:01:16 +0100
Hello,
And mira indeed thought that the memory requirements would explode, it
stopped.
"Loading skim hits from disk (will need approximately 156.0 GiB RAM)
...Ouch, out of memory detected."
Is there a setting I can use, so I can assemble this thing ?
With kind regards,
Jan
On Sat, Feb 28, 2009 at 22:20, Jan van Haarst <jan@xxxxxxxxxxxxx> wrote:
> Hello Bastien,
>
> I just started the assembly with "-SKIM:max_megahub_ratio=1" , and I'll
> track the memory usage.
> I'll let you know how it goes.
>
> Right now we aren't sure of the size of this genome, Newbler and Celera
> Assembler come to differing sizes, that's why I'm trying Mira to give me the
> definitive answer :-)
>
> Dag,
> Jan
>
> On Sat, Feb 28, 2009 at 21:59, Bastien Chevreux <bach@xxxxxxxxxxxx> wrote:
>
>> On Saturday 28 February 2009 Jan van Haarst wrote:
>> > What should I do to get the assembly to finish ?
>> > 0.5 % of bad reads doesn't sound that bad, does it ?
>>
>> Hello Jan,
>>
>> no, it doesn't. I suppose that the megahubs detected come more from the
>> fact
>> you have an eukaryote than from remaining vector adaptor sequences.
>>
>> Set the allowed megahub ratio to 1.0, that should allow you to continue.
>>
>> If I did my math right, you seem to have something like a 40 to 45 MB
>> organism
>> ... or 70 to 80MB if the main bulk of your reads is Titanium instead of
>> FLX.
>> Please note that I had mixed feedback from users regarding eukaryotes of
>> this
>> size, seems like MIRA sometimes explodes in memory consumption for these
>> (I'm
>> working on it).
>>
>> 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
>>
>
>
>
> --
> Dag,
> Jan
>
--
Dag,
Jan
Other related posts:
