Re: [Rqtl-disc] Alleles potentially switched at markers

[Karl Broman]

Typically it would be that the genotypes of the parental lines are different than believed at some portion of those markers.

At a SNP, say, you'd thought that line A was CC and line B was GG, but in reality it was the other way around.

karl


On Oct 9, 2012, at 9:42 AM, Starryfall <yiex...@gmail.com> wrote:

> Hi,
>
> Glad to find this group! I just started to learn QTL and I am using Dr.Broman's book "A Guide to QTL Mapping with R/qtl". I recently run into a problem. I performed est.rf on a set of data and got a warning message: Alleles potentially switched at markers. 35 markers jumped out after checkAlleles. There are some explanation on the book but I cannot understand it. Especially, how could this happen? The data was exported out from Fluidigm but not manually typed in. How does "alleles potentially switched at markers" happen?
>
> Thanks!
>
> --
> You received this message because you are subscribed to the Google Groups "R/qtl discussion" group.
> To view this discussion on the web visit https://groups.google.com/d/msg/rqtl-disc/-/gFo9k3W1wCAJ.
> To post to this group, send email to rqtl...@googlegroups.com.
> To unsubscribe from this group, send email to rqtl-disc+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/rqtl-disc?hl=en.

[Starryfall]

Thanks Dr. Broman!
I am not sure I understand~

The data I have is a F2 intercross from B6 and B10 mice.  The mutated male B6 was crossed to female B10 (6 sets of paired mating) and the offsprings were intercrossed to get F2 for SNP screening.  When we detect SNPs on the F2s, one pure B6, one pure B10 and one F1 were also run the same time as standards.  For each maker, based on the result from B6, we define B6 allele in all the F2s. eg: we name allele "X", "Y".  If for marker 1, B6 mouse show "XX", then all the other mice have "XX" on this marker is defined to contain B6 allelle.
Back to your reply, since the parental lines are F1s, offspring from B6 and B10 cross, their genotypes for SNPs should all be hets (B6/B10, XY). The F2s can be XX, XY, and YY.  Does the "Alleles potentially switched at markers" means that if the current allele show XX, it is actually YY?  What if the current allele is XY?  Is this purely caused by genotyping error?  What is the best way to solve this, swap the alleles or re-run the SNPs?

Thanks

[Karl Broman]

If you look at two tightly linked markers, it should be that the most common two-locus genotypes are
XX:XX, XY:XY, and YY:YY.

The markers indicated to have potentially switched alleles have a large LOD score (indicating linkage) with another marker, but with estimated recombination fraction > 1/2 rather than < 1/2. With such a marker, you should find another tightly linked marker for which the most common two-locus genotypes are XX:YY, XY:XY, and YY:XX. With such a pair of markers, it must be that one of them has the X and Y alleles switched.

If you have many F2 offspring and typed just a few of the B6 and B10 parents, such a finding might indicate genotyping errors in the B6 and B10 parents.

karl

> To view this discussion on the web visit https://groups.google.com/d/msg/rqtl-disc/-/t4FV-IFJIh0J.

[Starryfall]

I got it!  Thanks Dr. Broman!