Molly Fitzpatrick recently suggested that perhaps 23andMe has a future as a dating service. While her proposal probably involves a bit too much incest for the average single, there’s another proposal I recently dreamed up that may be even more promising.
Here’s the 3 steps that could turn 23andMe into a turn-key OkCupid:
1. It’s been scientifically demonstrated that people prefer mates with highly dissimilar HLA types. The effect is tied to variations in body odor and directly correlates with who people find most sexually attractive.
3. Either 23andMe or an enterprising 3rd party developer could create a lightweight app that compares the HLA data from 23andMe and mines it for anti-correlation. You can at least check people on a case by case basis and see if for instance, you and your current partner have the genes for robust, magnetic compatibility.
I fully expect that this kind of screen would reject close relatives as “subjectively unappealing to you” and would also find most people who found each other attractive could be identified correctly too.
The real question is whether the incomplete HLA data available via 23andMe and the currently limited knowledge of HLA/oder data has enough power to discriminate and reject most of the people who you’re only so-so matches with. If someone starts working on this, let me know. Online dating is a wasteland of faux-innovation but this kind of “OkCupid meets 23andMe” idea is the kind of startup idea revolutionary enough that even I would invest in it.