Reply To: "Carolina dogs" in reference panel?

[jessica hekman]

Hi, Stephanie. I took a look at Macy’s results and they look reasonable to me, though I can completely see how they would be surprising. The 30% unknown may well be Carolina Dog. There’s another 30% “other”. The breeds that are in that list are stuck in there because they’re all at less than 5%, which means we don’t really trust that they’re for real. So you can just as well look at her as 60% unknown/possible CD. (It’s debatable whether we should include the breeds that we have low confidence on, but we opted to give folks more information back rather than less.)

For the rest, I’d feel quite confident that she does have a grandparent who was a purebred Collie, leaving her with about 25% Collie. But all in all, she is a mixy-mix, and when you have so many breeds in there, the traits can combine in some really surprising ways. So it isn’t too surprising that she doesn’t really look like one thing or another.

Let me know if you want to talk more about this, either here or by emailing team@darwinsark.org and asking them to transfer the mail to me.

Jessica
Darwin’s Ark Researcher