Trait A has the property that whether you have or don't have a particular expresion of it, you seek mates in the same condition.
Trait B has the property that you randomly select mates without regard to their trait B status matching yours.
Trait A would tend to persist or even be amplified throughout generations, while trait B would be dispersed and become less obvious over generations.
Given the obviousness of race during mate selection, and the fact that the human race as a whole has not homogenized in that dimension, I would say that there is some non-trivial natural preference among humans to pick "like" when seeking mates.
(There are of course obvious flaws in the above thinking, the most obvious I can think of being that societal mobility has only in the last few hundred years made it reasonably practical to date outside the local area in which you were born. I do however think that there's a genetic basis for "birds of a feather" in mate selection, and that that doesn't imply the root cause is an undesirable form of racism.)
Give civilization another couple thousand years, and this might change. But if you look at the results of the Implicit Association Test, and the continued popularity of ethnic conflict, you've gotta assume that evolution has a big lead on political correctness.
By the way, have you taken the implicit association test? I highly recommend it for those of us who aren't biased. It's often illuminating.
Since this is potentially a sensitive subject, I'll preface my question with the assertion that I'm not offended, just curious.
I have no such bias; why do you consider it a "strange mutation?"