Sophisticated design; new ideas to me. I suspect that the genetic component of ascertainment bias of “participating” vs “not participating” would be swamped by other genetic factors.
But I could well be wrong. Love to read opinions on the HN from members of the statistical genetics crowd ;-)