[mira_talk] Re: mmhr problem

On Thursday 20 November 2008 23:59, Clayton Coffman wrote:
> I am having a problem running a de novo 454 est assembly.  Essentially it
> tells me I have lots of megahubs, and I am trying to track down what these
> are.  To do that I am trying to force it to go ahead with the assembly by
> setting mmhr=1 but it always aborts anyways saying the ratio is greater
> than 0, even though I set it to max at 1.  I could be doing it wrong, heres
> what I do:
>
> mira  -fasta -project=Px -SK:mmhr=1  -job=denovo,est,draft,454

Hello Clayton,

I need to clarify in my docs that the quick switches (--job=... and friends) 
should be used towards the front of the command line as they overwrite almost 
every other option of MIRA.

So, use:  "mira  -fasta -project=Px  -job=denovo,est,draft,454 -SK:mmhr=1"

and you're good to go.

> Is there a better way to find out what the megahub is?  My sequences aren't
> paired-end and I set ssf_extract to trim an apporpriate number of bases on
> the left to account for an adapter which I know is supposed to be there.

To see whether the adaptor is consistently on the left side of your reads, use 
sff_extract once without a left clip. If the adaptor sequence is consistently 
there, sff_extract will report that.

The following is a short guide on how to find out the really nasty repeats in 
your reads. I admittedly need to smooth out a few things with MIRA, but at 
least it works :-)

Run the assembly in a separate directory once with 
   "-SK:mnr=yes:rt=<some-int-between-5-and-10>"

MIRA will then mask the <int>-% highest occuring k-mers in your reads and 
report these to a file in the log directory.

This happens almost directly after loading, so you can CTRL-C the program once 
you've seen these lines:


---------------------------------------------------------------------------

Skimming for repeats (1/3)
 [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... 
[40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... 
[80%] ....|.... [90%] ....|.... [100%]
Compressing hash histogram ... done. Sorting ... (this may take a while) ... 
done.
Used hashes: 6105868
Unused hashes: 262329588
Median hashes: 14
Alternative median hashes: 21
Max hashes: 265
Masking starts at: 210

Skimming for repeats (2/3)
 [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... 
[40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... 
[80%] ....|.... [90%] ....|.... [100%]

Skimming for repeats (3/3)
 [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... 
[40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... 
[80%] ....|.... [90%] ....|.... [100%]
Localtime: Fri Nov 21 00:48:02 2008

---------------------------------------------------------------------------

The file you should look for is 
named "*_int_skimmarknastyrepeats_nastyseq_preassembly.0.lst" and is in tab 
delimited format (name, masked sequence):

nGGMAW54TR     GGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGG
nGGMAY80TF     GGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGG
nGGMB067TR     GGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGG
nGGMBD71TR     GGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGGCTTCGG

Hope it helps,
  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: