[mira_talk] Re: haplotype phasing
- From: Bastien Chevreux <bach@xxxxxxxxxxxx>
- To: mira_talk@xxxxxxxxxxxxx
- Date: Fri, 31 Jan 2014 20:58:00 +0100
On 31 Jan 2014, at 19:57 , Adrian Pelin <apelin20@xxxxxxxxx> wrote:
> I was wonder if the contigs that MIRA build due to intragenomic
> variation/heterozygosity are phased haplotypes?
> In other words, could we expect these small contigs to represent portions of
> the chromatids?
Most of the time yes. At least if SNPs determining the haplotype are closer
together that the average read length (or the average library size in paired
libs).
But things are not always easy. Assume the following to be “the truth” in
reality:
haplotype 1: xxxxAxxxx … 1000 bp … yyyyAyyyy
haplotype 2: xxxxCxxxx … 1000 bp … yyyyCyyyy
Now, if you sequence that with, say, Illumina 100bp paired end, you may get the
following four cases (with equal chances):
I)
1 large contig with xxxxAxxxx … 1000 bp … yyyyAyyyy
+ 2 small contigs, one xxxxCxxxx and another one yyyyCyyyy
II)
1 large contig with xxxxAxxxx … 1000 bp … yyyyCyyyy
+ 2 small contigs, one xxxxCxxxx and another one yyyyAyyyy
III)
1 large contig with xxxxCxxxx … 1000 bp … yyyyAyyyy
+ 2 small contigs, one xxxxAxxxx and another one yyyyCyyyy
IV)
1 large contig with xxxxCxxxx … 1000 bp … yyyyCyyyy
+ 2 small contigs, one xxxxAxxxx and another one yyyyAyyyy
The reason for this being that the placeholder “1000 bp” stands for absolute
identical sequence which - in absence of further SNPs, longer reads or
libraries with larger insert sizes - make it impossible impossible to cross
correctly.
B.
--
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:
