Quartz has an article up, 23andMe has a problem when it comes to ancestry reports for people of color, which I want to comment on at length. Though literally taken the title is not something I’d disagree with too much, the tone and details I have serious issues with.
First, some disclosure. Hong talked to me on the phone for an hour about this story. Mostly we talked about her Korean ancestry results. More on that later. Second, I consulted for 2.5 years for Family Tree DNA, am friends with Spencer Wells (who is quoted), and am on friendly terms (I’d like to think!) with Joanna Mountain, and quite respect many of the scientists at 23andMe (e.g., Kaisa Bryc and Ivan Juric off the top of my head).
I will go through the article point by point. First:
I doubt that most 23andMe users realize how paltry the company’s data is for non-Caucasians. For example: The data set that 23andMe used to generate my report has 76 Koreans in it, according to Dr. Joanna Mountain, the company’s senior director of research. 76 Koreans. It is estimated there are at least 7 million Koreans living outside of the Korean peninsula—including 1.7 million in the US—among a worldwide population of 83 million.
Seventy-six Koreans seemed small to me, but what do I know? I’m just a journalist. So I spoke to geneticist Spencer Wells, founder and former director of National Geographic’s Genographic Project (arguably a 23andMe competitor), which he ran from 2005-2015. “ is a really low number,” he concurred.
The small sample sizes seem really, really problematic if you are a lay person, or a journalist. The issue is that with genotype technology that looks for common polymorphisms you really don’t get that much more information from 1,000 individuals than you do from 100. All things equal, more sample size is better, but the gap between 10 and 100 is much much greater than 100 and 1,000 or 100 and 10,000. You can see this in the robustness of results for model based clustering conditional on different sample sizes. For a homogeneous population like the peoples of the Korean peninsula, who seem relatively panmictic, a bigger sample size would have only marginal effect on the overall outcomes using these methods (also, it might matter if you were looking at low-frequency alleles from whole genome sequencing).
Before I talked to Hong I checked in with a friend who was half north Korean (in that her father’s family was from the northern half of the peninsula and migrated south) and half central Korean (i.e., her mother’s family was from around Seoul). Just like her husband, whose family was from Busan in the far south, her results came back as 99% Korean. Some genetic research has been done on Koreans, and there just isn’t that much structure. The Koreans have a composite origin if you go far back enough, but they’ve been intermarrying with each other a long time.
Also, astonishingly, the report shows that I am 13.4% Japanese and 14% Chinese—and only 61.6% Korean. I was looking forward to watching my parents freak out. My sister texted me, “Oh [Dad will] probably blame Mom.”
To my disappointment, my parents did not freak out, nor did they get into an amusing argument about which of their ancestors was the ho. Because they simply did not believe the data. And, for once, they were right.
The public relies on journalists for the truth. Sometimes the truth can be slippery. But sometimes it is clear. Most of conversation between Hong and myself was about her Korean ancestry. As I said to her, I asked a handful of my Korean friends about their 23andMe results before we spoke. From that I told Hong I was 99% sure that she had recent non-Korean ancestry. 23andMe’s results are really robust. I tried to emphasize that over and over. Hong can believe what she wants, but it is obvious that she almost certainly has non-Korean ancestry relatively recently in the past.
Because 23andMe uses chromosome painting, you can see she has very long segments of inferred Chinese and Japanese ancestry. This non-Korean ancestry is probably from within the last three generations because ancestry tract lengths indicate that recombination hasn’t broken apart the associations across the chromosomes (there are 20-40 recombination events across the genome per generation).
I asked Wells whether my percentage breakdowns of Korean, Chinese, and Japanese meant anything. “Yes,” he said, “but I think it is misleading to go to a decimal place or even to go out two digits.” Wells said that another problem with the data is that “Most of those [samples] are from the US. They’re not terribly useful for studies of indigenous composition—which is effectively what this analysis is trying to do.”
I had a long text conversation with Spencer on this after the article came out. I can see where he’s coming from. And 23andMe does have a shortfall of indigenous and non-European samples. But as I said, I asked around to Korean friends who had used 23andMe before and the population is pretty homogeneous, and the friends’ results I cited above were representative. I have also worked with and seen samples from Family Tree DNA, and it’s the same story. There might be undersampled populations from Korea, but I’d bet against it. Koreans are relatively homogeneous, with a position between Japanese and North Chinese. Where you would expect them to be.
Spencer is correct about the decimal places issue. They give people a false impression of precision. I do know that scientists within DTC companies struggle against it. But scientists don’t always win these arguments.
I also interviewed Harvard geneticist Robert Green, who made the important point that private companies have different methods and standards from those of an academic lab. “There is a difference between analysis you can do with hundreds of [genetic] markers at a research level, and the kind of analysis that even the best companies can do, which is more an approximation,” he said.
Green is a medical geneticist who does great work. But I’ll be generous and assume he’s taken totally out of context here, because what he says makes no sense. The genotyping platforms do have error rates (no-calls, mistypings, etc.) on the order of 1%. But they’re using hundreds of thousands of SNPs. This error rate doesn’t matter too much for what 23andMe is doing in relation to ancestry. And with population structure inference these errors usually don’t cause a major issue if they aren’t systematic.
Then there’s this:
A few of the geneticists I interviewed for this article (but not Green or Wells) outright accused 23andMe of commercially driven ethnic bias. For example, no distinction is made between northern and southern Chinese, who have very different traits. This was a serious allegation, so I put the question directly before 23andMe’s Mountain. “As a scientist, I find that insulting,” she said in a phone interview.
I brought up the issue with the Chinese to Hong, and I apologize to Mountain here if it came off as offensive, because I certainly didn’t mean it that way. My point, which I’ve brought up for years both in public, and when I have consulted for DTC companies, is that South and East Asians are huge groups, and it’s incongruous that they aren’t differentiated as much as the Europeans. These tests basically tell you are South Asian, or Chinese, or Korean, or Japanese. In the case of Koreans and Japanese there isn’t that much structure within these groups, but that is not the case with the Han Chinese. There is an decent amount of structure, but last I checked 23andMe has a catchall Han Chinese group. Why? I’ll get to that later. (It’s not because they don’t have the data.)
Though I disagree with the tone and the emphasis, a simple inspection by Hong has shed light on something that has been glaringly obvious in the genetic genealogy community: there is laser-like focus on differentiating very close Northern European groups, such as Irish and English, and not so much emphasis on differentiating diverse populations such as South Asians. This was one thing I did talk to Hong about at length. I don’t think it’s crass racism, and I think that I made that clear to her, but I’m not happy with the situation either (23andMe representatives know I’m not happy, and have talked to me about it at ASHG).
The final sections involve Hong reviewing the disparities in sample representation. As I said above, some of this overdone. But, it is a little ridiculous that there are only a few hundred African population samples in their data. Granted, it turns out that between-population genetic distance in Africa is actually not as much as you’d think based on aggregate variation (the within population variation is what makes all the news). I think Hong is correct that 23andMe should have made more effort on sample collection these past few years…but I’m not CEO of 23andMe, and Joanna Mountain and her scientists don’t call all the shots. I think Hong’s piece leaves Mountain and the researchers holding the bag for something that really isn’t their doing (perhaps it is, but I’m really skeptical of that).
Could the company be doing a better job with collecting ethnographic data? “Absolutely they could,” Wells said, “but it’s not their raison d’être.” Which, of course, is pharma and health research. Fair enough—it’s their money. But how about a disclaimer attached to the ancestry part of the report? Like, “for entertainment purposes only?” Because data based on 76 Koreans (or any other ethnic group) is definitely not worth potentially causing family discord or a blood feud. I don’t know whether the company understands the realities of deadly global ethnic tensions and the potential damage created by people’s trust in these reports.
I think Spencer has highlighted the major dynamic here: 23andMe is pivoting towards biomedical research. It has a database of north of a million, mostly European-origin individuals. The real money now comes from leveraging the database to collect information on health, and combining it with the genotypes they already have. On the margin, getting greater population diversity is probably not a major avenue by which they could gain higher valuations. And getting from one million to ten million genotypes is nothing without increasing their database of phenotypes.
The real story here is not one of racism. It’s one of capitalism. Most of 23andMe’s customers are white European in ancestry, and a disproportionate number of those are Northern European. Is it a surprise that their tools breakdown Northern European ancestry so finely? That’s their customer base.
Second, many Asians I’ve talked to are relatively uninterested in fine-grained breakdowns in their ancestry. For several years I worked with an engineer from Fujian, and his Family Tree DNA results showed that he was shifted toward the southern end of the north-south Chinese cline. He didn’t care at all, because he was from Fujian, so of course he knew this. Many Asians seem to have this attitude where the ancestry results are viewed as confirmatory. Hong’s case, where there was a surprise, is exceptional.
If 23andMe wanted to they could easily breakdown Asians into further subcomponents. I think there are two reasons they don’t want to aside from the firm’s recent focus on health and pharma. First, they don’t have that many Asian customers. Second, their Asian customers might actually get a bit irritated!
Ultimately, Hong can think whatever she wants to about her 23andMe results. But the data are out there. It’s pretty obvious that unless there was a sample mix-up, she has recent Chinese and Japanese ancestry (she could put the raw results in the public domain and have people cross-check with other methods, like PCA; I’m pretty sure they would confirm the 23andMe results).
On a last nerdy note: the data generated by DTC companies is great. Their Illumina SNP-chips are really good, with 99% or so correct-call rates. Hong referred to data in the piece when she really meant results. The thing is that results are basically generated through a sieve of methods geared toward human digestibility. 23andMe and other DTC companies differ because of different methods and parameters in those methods, that are determined by what humans want out of these techniques. But the data, that’s pretty straightforward and robust.
If you are interested in a more philosophical take, Joe Pickrell’s What is ancestry?
Addendum: My conversation with Hong was very wide-ranging. We talked about EDAR, random mating populations, and local ancestry deconvolution. Well, perhaps not in those words. It’s a little saddening to me that ultimately what came out of all that is a piece which tries to paint 23andMe as prejudiced against minorities. The only prejudice they exhibit as a firm is against smaller market share.
No matter the Yelp reviews, if it doesn’t have dry pot or whole boiled fish on the menu, not worth it. Also, should feature something where the peppercorn is salient.
What a disappointment. Salty. Without much other flavor besides the spice. It was like a watery spin on Louisiana hot sauce. I couldn’t taste the “aromatic spices” and “fresh herbs.” And don’t tell me it is because it’s too spicy, I didn’t find it too spicy. I did find it very salty though.
In 2011 I was having dinner with an old friend who was an engineer at Intel. He also has a Ph.D. from MIT. Smart guy. But when I mentioned casually offhand that we were all a few percent Neanderthal (outside of Africa), he was surprised. I was a bit shocked, as I explained that this was a huge science story. The Neanderthal genome had been published the previous year. How could my friend not have known?
He was totally unembarrassed, and told me I overestimated how closely the public followed genetics and paleontology. I’m sure he was right. But it’s hard to remember sometimes.
We’ve gone further beyond where we were in 2010. We now have a really good grasp of a lot of population dynamics in Eurasia over the past 20,000 years. Probably the best place to start is with this preprint, The genetic structure of the world’s first farmers. But the general outlines were already evident a few years back in Toward a new history and geography of human genes informed by ancient DNA.
Most of the world’s population seems to descend from a mixing of a set of groups which 10,000 years ago were distinct. How distinct? We’re talking about Fst values on the order of 0.10, which means that ~10% of the variation genetically is partitioned across two pairwise populations. That’s about what you see between Europeans and Chinese today. Some of the Fst values were a bit higher, some lower, but the 0.10 seems about right.
To make it easy for some of you, I’ve labeled and placed the approximate locations of ancestral groups to modern Northern Europeans ~10,000 years ago. What I’m trying to represent is a map which shows the modal regions of distribution of ancestors that Northern Europeans today had 10,000 years ago. So, for example, since ~15% of the ancestry of Northern Europeans is “Ancient North Eurasian” (ANE), a lot of ancestors of Northern Europeans alive today would be living somewhere in the broad expanse of Central Eurasia (now, because of various demographic events the number of ANE was probably lower than farmers, perhaps lower than the 15% contribution to the modern genomes).
A substantial proportion of the ancestry of Northern Europeans is “European hunter-gatherer,” dating to the Pleistocene. But here’s the kicker: most of that ancestry dates to after the LGM, to about ~15,000 years ago. The really deep Pleistocene ancestry in Europe is only found at very low levels now.
The final issue is that a lot of the phenotypes that we racially code are recent. This probably explains why groups like the Kalash and Nuristanis can look more like Europeans than South Asians, but they’re genetically more like South Asians.
What does any of this have to do with non-scientific things? I don’t really know. My interest in population structure is intellectual, not personal. But a certain type of person should probably stop talking about how white people have been in Europe for 40,000 years. First, the ancestors of modern Europeans 40,000 years ago were almost all residing outside of Europe. An assertion that holds until 15,000 years ago. And most would still be resident outside of Europe 8,000 years ago as depending on how you count/calculate* And, perhaps more importantly, the typical phenotype of Northern Europeans probably really coalesced only around ~5,000 years ago.
* Definitely true for Southern Europeans, but conditional on Northern Europeans depending on where you draw Europe’s eastern boundary.
Addendum: I stole the title from John McWhorter’s book, Our Magnificent Bastard Tongue.
Also, this is not to say that
1) population structure today is trivial in a phylogenetic sense, it isn’t.
2) it is not to say that population structure functionally irrelevant, it isn’t.
So I don’t really have strong opinions on the whole controversy over women’s sports at the elite level…mostly because I have a really hard time following all the logic. For me the biggest problem seems to be that we have two categories, men’s and women’s, and there are those who are arguing that they’re actually nearly plastic catchalls…which then suggests to me we shouldn’t have two categories in the first place in competition at the highest levels.
With that in mind, D. J. Grothe points me to this prescient interview from a few months back, Hyperandrogenism and women vs women vs men in sport: A Q&A with Joanna Harper. Joanna Harper is a transwoman who is (was?) also a competitive racer and a sports scientist. This portion is where the facts stand:
I would also like to relate a two-part epiphany that I had after my transition. In 2005, nine months after starting HRT, I was running 12% slower than I had run with male T levels; women run 10-12% slower than men over a wide range of distances. In 2006 I met another trans woman runner and the she had the same experience. I later discovered that, if aging is factored in, this 10-12% loss of speed is standard among trans women endurance athletes. The realization that one can take a male distance runner, make that runner hormonally female, and wind up with a female distance runner of the same relative capability was life changing for me.
As they say, “read the whole thing.” It’s long, and detailed, and doesn’t offer easy answers. Ultimately the reality is that no “solution” is going to be fair to world-class athletes. But, it’s probably important to remind ourselves that it is also unfair to those of us without the genetics of world-class athletes, and we seem to be OK with that.
Compare and contrast with this piece from Let Caster Run! We Should Celebrate Semenya’s Extraordinary Talent. The title really captures the reality that it was pretty obvious that the author was going to come down on one side, and would make a lawyerly case. Rather disappointed with Nate Silver’s shop.
Guy Gavriel Kay’s Children of Earth and Sky is set in the same world as the Sarantine Mosaic duology, and the Lions of Al-Rassan, The Last Light of the Sun, and A Song for Arbonne. I’ve enjoyed Kay’s work for more than half my life at this point, so no surprise that I enjoy Children of Earth and Sky. As I’ve noted before, Kay is arguably the world’s greatest historical fantasist, and for someone like me it’s always pleasurable to make connections between our own real history, and his secondary creation. This sort of fantasy is more magical, than characterized by magic.
I know I have readers in India because of IP addresses. Keep an eye out for my byline in India Today, where I’ll make some contributions now and then. The first should drop this Friday in print and online, a short review of Shadi Hamid’s Islamic Exceptionalism. An Indian friend told me that India is one nation where the sales of print are actually increasing, so I’m curious how this will go.
On Twitter most of my blocks come because I’m being tweeted at directly by someone. If I don’t follow you on Twitter in most cases I don’t want to be bothered. The main reason I block isn’t because I’m a coward or I feel unsafe. It’s because the person is probably stupid, and starting to annoy me. Sometimes, it’s because they want me to make a point that they want to make. Needless to say, I don’t take kindly to that. Between all my various adult responsibilities that I have now at this age, I don’t really feel guilty at all muting stupid people (who invariably think they’re genius, because you know, they’re stupid).
Also, Running Structure-like Population Genetic Analyses with R. Looks like there are some interesting visualizations of admixture components which are feasible with the new program.
People keep emailing me about the HGDP plink data set. I think I removed where it initially was, and it’s linked to my old Admixture tutorials. Well, download this zip, and look at the .fam file. It has clear population labels, so you should be able to do what you want in Plink.
Should We Be Having Kids In The Age Of Climate Change? The arguments really go all the way back to the ZPG movement. Actually, they’ve popped up in philosophical movements from the beginning of time. The world is a “vale of tears”, etc. Myself, I have no guilt about having children. My children are attractive, and seem rather intelligent so far. Would that more children like mine exist in the world!
Stop Tweeting Your #Firstsevenjobs: It’s just a way to disguise your privilege. FUCK YOU. The author of the piece has a degree in French language and literature from Columbia university according to her Linkedin. She gets to write for a living for Slate about food, and was editorial assistant for Mark Bittman. What. The. Fuck. She gets paid to write about food! She was Mark Bittman’s assistant. I guess it takes one to know one. Not that the author tries to mask her privilege: “But this list doesn’t tell you that I went to an Ivy League school and graduated without debt, since my parents were able and willing to pay for my tuition.”
I hate it when people say that gender is a continuum, because that tacitly brings to mind a uniform distribution. It’s not. It’s highly bimodal.
Lou Pearlman is dead. The weirdest thing about his career is that several stories have implied he became a boy-band impresario because he was a closeted gay man, and that was a way for him to have access to young vulnerable teenagers. The fact that he became very prominent in the late 1990s boy-band boom was almost a coincidence.
Sarah Haider has been accused of being a white supremacist.
Let Caster Run! We Should Celebrate Semenya’s Extraordinary Talent. As they would say, “I don’t even.”
The company I work for has a 20% discount on kits right now. So if you want your dog to get genotyped on 200,000 markers, and get ancestry and health, it’s a good time as any.
What are you reading?
The Amazing Atheist youtube channel has some pretty funny videos. E.g.,
Taking a break in my work of the day I stumbled upon the fact that Bernard Cornwell’s series based on King Alfred’s period, which began with The Last Kingdom, is a Netflix series. To be honest I much preferred the three volume Warlord Chronicles, set more than three centuries earlier, in post-Roman and pre-Saxon Britain. A retelling of the Arthurian romance with not too much romance, George R. R. Martin admitted to me in correspondence in the late 1990s that he quite enjoyed it as well. The protagonist of The Last Kingdom is peculiarly similar to the one in Warlord Chronicles.
As a fan of alternate history I’ve occasionally stumbled upon the “what-if” scenario whereby Alfred’s Wessex is conquered, and England becomes Daneland. Would we today be speaking another Scandinavian language? Would Christianity disappear, and the pagan rites of the Norse come to rule the day? It seems broadly likely that that would not be cause at all.
First, the victory of Christianity in Europe was overdetermined by the 9th century. Even in this period there was a Christian presence in Scandinavia. A Scandinavian ruled England would almost certainly be a Christian one. And in fact in the century before the Norman conquest the Scandinavians created a hybrid society with the native English. Harold Godwinson had a Danish mother, and connections to the Danish monarchy.
The second issue is one of language. The English language of Alfred’s time was much more Germanic, so the gap between it and the tongue of the Danes was not that large in any case. And, from what I have seen, it seems that the number of Scandinavians in relation to the native population was much smaller than that of the Saxons in relation to the British, though even in the latter case it must be acknowledged that the Germans who arrived in the 5th to 6th centuries were numerically outnumbered by the native Romano-British (see PoBI results).
Perhaps if the kingdom of Wessex fell England’s identity would be more indubitably aligned with Scandinavia, as it was arguably in the decades before Norman conquest in any case. But cultural identities can be curiously resilient. The Finns endured nearly 600 years of Scandinavian domination, but maintained their language, while the long Irish interaction with the Vikings still left the Irish identity intact.
They promote their new project, seeq. It looks pretty slick, and I’m excited to be part of the batch of beta testers.
Obviously I’m doing more development right now than I would have expected. But in the long term I want to move beyond hacking to survive for the present, and write some code that’s sustainable. So I think I want to read a design patterns book. The last one I read was 15 years ago and I don’t really have much retention of it. I’m particularly interested in stuff geared toward Python (the language I’m starting to get comfortable in right now).
Readers with recommendations are invited to weigh in. I know I have a fair number of software engineers in the readership, so I’m asking for your thoughts and suggestions. Perhaps the classic from the GoF is still the way to go? Remember, I’m not a software engineering who works on scientific data, I’m a scientists who sometimes needs to do a little engineering and data analysis.