[mira_talk] Re: Solexa paired-end reads
- From: Peter <peter@xxxxxxxxxxxxxxxxxxxxx>
- To: a.travis@xxxxxxxxxx
- Date: Thu, 24 Feb 2011 12:08:57 +0000
Hi Tony,
On Thu, Feb 24, 2011 at 11:46 AM, Tony Travis <a.travis@xxxxxxxxxx> wrote:
>
> Hi, Peter.
>
> We use Hawkeye to look for misassemblies, where mate-pairs are in AMOS
> nomenclature 'unhappy' (i.e. stretched or compressed in comparison to the
> expected insert length). It would be very useful to have a similar function
> in Tablet, and I've already mentioned comparison with Hawkeye to Iain Milne.
> I think Tablet is getting more useful as it develops, but Hawkeye is popular
> for a good reason: Detection of misassemblies.
I believe (having not really used Hawkeye first hand), that it colours the
paired end reads based on the fragment length (insert size). At least I
vaguely recall you showing screenshots illustrating this in a NextGenBUG
talk.
I've suggested such a fragment length based read colour scheme to the
Tablet guys - it think it could work quite well even without having to tell
Tablet the expected fragment size (although that is something that could
be recorded in the SAM/BAM header I think, not sure about other
alignment formats).
A different solution is to exploit the fact the SAM/BAM format allows for
paired reads to be marked as a "nice pair" or not (FLAG field bit 0x02,
each fragment properly aligned according to the aligner). Tablet might be
able to show that status visually somehow. Then the question becomes
how to set that bit when generating the SAM file from MAF... perhaps
Bastien has some ideas?
> The problem with Hawkeye is that it is difficult to install and run. That,
> in my opinion, is a major advantage of Tablet over Hawkeye but without
> mis-assembly metrics, Tablet is not as useful for my work. Despite this,
> I'm now starting to use Tablet because it's good :-)
What is does, Tablet does quite nicely - but I think everyone has a
different list of priorities for missing functionality, I hope you are giving
them direct feedback (something I do).
Peter
--
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:
