While I can not bring up the research paper from memory, there is such evidence but it more funnier than that. Going directly by vague recollection of the lecture, there is a bias towards X distances of relativeness, something like 3rd or 4th cousin, where you get a statistical bias in a population. You also get this in animal studies. If I recall correctly it was part of when they talked about gene selection theory, and also (very vaguely recalled) a short aside into game theory for maximized gene transmission with minimal risk for genetic errors.

Technically this has nothing to do with racial and ethnic groups, but as an outcome it correlates.

(Edit: Remembering an other detail. The human study was a olfaction test where participants felt more attracted to the sweat of said relativeness. It add to the (in my view) comical aspect of it all).