Re: [svtoolkit-help] Overlapping CNVs?

[Bob H. <han...@br...>]

Hi, Eddie,

First of all, the examples you are showing below are not called as 
deletions - all three failed both the COVERAGE and DEPTH default filters.
The VCF will list all of the sites evaluated, but most evaluated sites 
are not in fact polymorphic.

That said, you will likely find some examples of overlapping calls among 
the passing calls as well.  In my experience, the number of
passing calls that overlap are usually a percent or two.  When there are 
overlapping calls, they _might_ be two different events, with different
breakpoints, but in most cases they are usually redundant calls of the 
same underlying polymorphism with slightly different boundary estimates.

I have been experimenting with filters to remove the redundant calls 
after genotyping.
These tools have been used in the 1000 genomes phase 1 processing, but 
are not released yet.
In the meantime, don't be surprised if you get a few redundant calls.

-Bob

On 4/10/12 4:48 PM, Eddie Loh wrote:
> Hi,
>
> I ran Genome-STRiP on my dataset and in the output genotype vcf files, I
> noticed that some of the reported deletions were overlapping. How should I
> interpret such overlapping deletions? Would one of the predicted deletion
> be better than the others?
>
> For example:
> chr1    2583913 DEL_13  C<DEL>    .       COVERAGE;DEPTH
> CIEND=-44,45;CIPOS=-44,45;END=2629140;GSCOHERENCE=-6.6125886653079;GSCOHFN=-2.2041962217693;GSCOHPVALUE=0.0399;GSCOORDS=2583629,2583912,2629141,2629308;GSDEPTHCALLTHRESHOLD=1.8113119220635596;GSDEPTHNOBSSAMPLES=2;GSDEPTHNTOTALSAMPLES=2;GSDEPTHOBSSAMPLES=0904,0906;GSDEPTHPVALUE=0.0;GSDEPTHRANKSUMPVALUE=0.913914;GSDEPTHRATIO=1.3544900147632686;GSDMAX=45328;GSDMIN=44329;GSDOPT=45140;<rest
> of line clipped>
> chr1    2583923 DEL_7   A<DEL>    .       COVERAGE;DEPTH
> CIEND=-59,59;CIPOS=-59,59;END=2621941;GSCOHERENCE=-3.731442108152636;GSCOHFN=-0.7462884216305272;GSCOHPVALUE=0.6772;GSCOORDS=2583656,2583912,2621951,2622152;GSDEPTHCALLTHRESHOLD=1.6648659343192074;GSDEPTHNOBSSAMPLES=3;GSDEPTHNTOTALSAMPLES=3;GSDEPTHOBSSAMPLES=0708,0904,0913;GSDEPTHPVALUE=0.0;GSDEPTHRANKSUMPVALUE=0.8289357;GSDEPTHRATIO=1.2133095867419799;GSDMAX=38138;GSDMIN=37139;GSDOPT=37940;<rest
> of line clipped>
> chr1    2583951 DEL_9   A<DEL>    .       COVERAGE;DEPTH
> CIEND=-121,121;CIPOS=-121,121;END=2617214;GSCOHERENCE=-1.3613655977544998;GSCOHFN=-0.6806827988772499;GSCOHPVALUE=0.601;GSCOORDS=2583824,2583939,2617225,2617334;GSDEPTHCALLTHRESHOLD=1.6374681994307405;GSDEPTHNOBSSAMPLES=2;GSDEPTHNTOTALSAMPLES=2;GSDEPTHOBSSAMPLES=0904,0912;GSDEPTHPVALUE=0.0;GSDEPTHRANKSUMPVALUE=0.9686783;GSDEPTHRATIO=1.4466964842117855;GSDMAX=33385;GSDMIN=32386;GSDOPT=33065;<rest
> of line clipped>
>
> Cheers,
> Eddie
>
>
>
>
> ------------------------------------------------------------------------------
> Better than sec? Nothing is better than sec when it comes to
> monitoring Big Data applications. Try Boundary one-second
> resolution app monitoring today. Free.
> http://p.sf.net/sfu/Boundary-dev2dev
> _______________________________________________
> svtoolkit-help mailing list
> svt...@li...
> https://lists.sourceforge.net/lists/listinfo/svtoolkit-help