Ed Yong has a post up on a behavior genetic publication where the sample size is 23. The researchers report a correlation between a SNP on the OXTR locus and “prosociality.” To make a long story short the sample size suggested to Dr. Daniel MacArthur and Dr. Jospeh Pickrell that this was a spurious correlation. The bigger issue here is that there are functional reasons to assume that some genes are correlated with normal human variation in psychology and behavior, and a robust body of literature that these traits are heritable (trait value is highly predictive across relatives), but, the results associating a particular genetic marker with a given trait are much less robust.

But I immediately realized something interesting: a sample size of 23 may be small, but there is a sample size potentially of thousands! I know my genotype at this SNP from 23andMe. How about some 23andMe customers get together and produce some results, and then get published in PNAS? A sample size of 230 would be easy I think, and you could probably push it much closer to 1,000.

Discover has added a new blog, The Crux. It’s a group weblog, and I’ll contribute now and then.

In a rambling column at Slate on (ir)religious intermarriage Jesse Berring observes:

Still, I concede that the irreducible alchemy of romance makes my cold logic rather difficult to apply to individual marriages. There are more things to a person—and to a relationship, one hopes—than religious beliefs. But since atheistic bachelors and bachelorettes are very rare specimens (there are no exact statistics available, but just 1.6 percent of the U.S. population self-identifies as “atheist”), deciding just how important it is to find a godless mate is indeed a real issue.

There are two small issues, and a big one. First, the 1.6 percent figure is a low one because the term “atheist” is somewhat taboo. Atheist as defined by those who don’t believe in God (as opposed to those who admit to being atheists) is closer to 5%. Second, the main issue with “atheist dating” is the sex ratio problem, though that’s more modest in younger age cohorts today than older ones.

But the broader point is that it’s totally ridiculous to assume that mating is random within the population. Jews are ~3% of the American population, but Jewish-Jewish pairings are not 0.03 × 0.03. I’m sure Bering saw that “religious nones” (of which 1/3 are de facto atheists) have a 50% probability of being with someone of the same lack of beliefs, despite being 15% of the population.

Overall I think it’s right that people should align reasonably well on big metaphysical questions for increased probability of amity. If possible. I don’t think metaphysics (or lack thereof) really matters much day to day, but it does start to matter when there’s a discordance. I just don’t get why Bering ends up writing stuff that’s plainly meant to provoke when there are serious and interesting questions which he is really addressing.

A few years ago I wondered offhand why Eurasians weren’t very pale, since East Asians and Europeans developed light skin at different loci over the past few tens of thousands of years. In hindsight the answer seems pretty obvious. I realized the solution when looking at the skin pigmentation loci in my parents’ genotypes. They’re both homozygous for the derived “light” variant of SLC24A5, but interestingly my father has more “light” alleles than my mother. This is peculiar because my mother is notably lighter complected than my father. Then I realized that there was a likelihood that my mother carried an East Asian allele which conferred light complexion, since she’s ~15% East Asian. So of course the reason that East Asian-European hybrids aren’t exceedingly pale is that pigmentation is predominantly additive in trait value effect and they’d be heterozygotes at many loci where their parental populations would be homozygotes.

On the other hand, the F2 generation might be potentially very light indeed (or dark)….

23andMe has a feature which allows you to check MHC compatibility. This is important for matching potential organ donors with those who need those organs. If a close family member is not a match (it’s a very polymorphic set of genes), then a co-ethnic is the next best bet. This is a major problem for those from ethnic minorities and of mixed-race. One day we’ll be able to “grow” organs from our own tissues, but until that day comes matching is still essential. Right now matching is done via drives and what not. But as more and more people get genotyped or sequenced, the information will be right there in the databases. Even racially mixed individuals with very rare combinations of alleles might find matches once the majority of the world’s population gets typed (though I haven’t done the math on this, has anyone?)!

Update: Just to be clear, as as Alex Khomenko observes this isn’t a prime time feature yet. It should be though in a few years. He also observes:

On the other hand, imagine the ethical implications of an accidentally discovered unwilling match being pressured to donate.

From what I recall about half the people who end up being matched decline to be organ donors.* I think that expanding the pool of potential matches so that this situation is minimized far outweighs the cost of the risk Alex alludes to above.

* Donating is not trivial. Going to a drive to check for a tissue match is. So it shouldn’t be that surprising.

Things we said today:

But change is afoot. Numerous teams of clinicians and genomicists (including two at my own institution) have come together to sequence patients’ genomes and/or exomes to identify disease-causing mutations.

Of course, doctors and researchers and genetic counselors are still bickering about when to sequence, whom to sequence, return of results, institutional liability, whether we are confusing research with patient care, and on and on. But for the moment, what everyone can agree on is that parents of kids with serious undiagnosed conditions likely to be genetic absolutely do not give a shit about any of those things.

They want help. They want answers. For two decades we have painted grandiose pictures of personalized medicine. Are we going to keep moving the goalposts? Are we going to tell them that we didn’t mean it?

There’s a chicken & egg problem with personal genomics. Until it gets ubiquitous and relatively transparent there is always going to be some letdown in terms of what it can, or can’t, tell us. But as long as there’s a letdown, it won’t become ubiquitous and transparent. That’s the main reason I put my genotype into the public domain. It doesn’t do me too much good or bad, but I wanted to show people that there’s nothing terrifying about it. Once the fear is gone, we can move forward and get some work done.

I haven’t had time to read a book front to back in 2 months. Probably the longest period I’ve gone like this since I was 13. I plan to “binge” as much as I can over the Holidays. Is there anything interesting you’re reading? And yes, I already have The Better Angels of Our Nature: Why Violence Has Declined on my Kindle.

Nice. We’re funded!:

Thanks to an amazing piece at the CNN blog Light Years by Ed Yong, the outpouring of support for the Roman DNA Project today has been astounding!  In financial news, we have actually exceeded our $6,000 goal, after just 10 days.  That goal was to fund analysis of at least 20 individuals (the immigrants to Rome that I found through Sr/O isotope analysis).  Of course, we are accepting donations through mid-December, so additional funding will be put to good use – studying more ancient Romans!

And I’ve received a dozen or more emails today from people as excited as I am about this project, offering their encouragement, lab services, expertise, and knowledge about the ancient world.  I will respond to all of them, I promise, but it might take a few days!

Again, thank you – all of you reading this – for making this project a reality!

Dienekes has a little extra commentary. In any case, congratulations are in order! This sort of genetic science seems to be “low hanging fruit.” It isn’t as if you need to fund a massive particle accelerator and what not. I hope that all the museums with assorted samples in their back rooms may now consider the possibility that they don’t have to wait for Max Planck to approach them….

I’ll Have the Red: Hot Sauce, Island by Island. Seems like the scotch bonnet is the basis of a lot of Caribbean sauces. You might be curious about my reviews of hot sauces from 4 years ago.

There’s a variable in the GSS, GENEEXPS, which asks if genes play a role in personality. The options are:

– It’s genes which play a major role

– It’s experience which determines personality

First, let’s admit that the premise is stupid. Personality is heritable, but environmental variation also seems to matter. In other words it is noncontroversial to assert that both genes and environment can explain variation in personality (or perhaps more precisely genetic variation can only explain around half the variation for any given trait).

I was curious how this broke down by education and intelligence. To remove demographic confounds I limited the sample to non-Hispanic whites. For intelligence I used WORDSUM, with scores 0-4 being dumb, 5-7 being average, and 8-10 being smart.

Read More