[mira_talk] Re: extremely high illumina coverage

  • From: Nestor Zaburannyi <nestor@xxxxxxxxxxxxx>
  • To: Bastien Chevreux <mira_talk@xxxxxxxxxxxxx>
  • Date: Tue, 18 Oct 2011 22:27:08 +0200

Dear Bastien,

Thanks  for  the answer. And if it is GC-rich? I am reluctant of using
more  than  100x,  because of people here saying it is not good. Also,
because MIRA gets very slow.

N.

> On Oct 18, 2011, at 22:06 , Nestor Zaburannyi wrote:
>> Can  i instruct MIRA to trash a read that it is going to map next if
>> the  coverage  is  more than n-fold in this particular place?

> No.

>> Can this feature be implemented easily?

> Yes and no. Yes because it would be a couple of lines, very easy.
> No, I won't do that because the effect would be totally different
> that you would expect: it would lead to a stall of contig
> construction and hence the assembly being fragment in thousands of pieces.

>> The reason for this - i have a Bacteria
>> with  800x  coverage.  When  i take a reasonable subset (80x), it gets
>> assembled,  but not so good. I *suspect* that with such approach i can
>> close many gaps with this coverage. Can MIRA community coment on this?

> If this is not a GC rich bug ... already with 80x I suspect that
> most things should assemble pretty well. You can give it a try at
> 120x or 160x, but if that does not assemble neither, you either

> - have a repeat there
> - caught a case where Illumina has severe sequencing problem
> - caught a case where MIRA is doing something foolish.

> B.



-- 
З повагою,
 Nestor                            mailto:nestor@xxxxxxxxxxxxx


--
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: