Credit: Robert Payne

A concern about the breach of privacy emerged almost immediately. Though I have serious reservations about the sensationalism which BritainsDNA has engaged in, I think it is totally legitimate of them to infer William’s ancestry in the fashion they did. First, Prince William is a public person, and in direct line to the throne of the United Kingdom. Though some of the spin may be distasteful, remember that this is a person who is where he is because of his ancestry. Second, anyone who performs genealogical research is exposing the information of family members, often without their consent. If William’s mtDNA haplogroup was known to be pathogenic than the case for withholding the information from the public seems straightforward. As it is all that was uncovered was relatively banal, that William may have a South Asian ancestress. There’s a lot of information about me that I’d rather not others know first, but that’s not how the world works. In the grand scheme of things this just isn’t a big deal, and we should focus on the more concrete problem of public understanding of science, and long term issues in regards to genetic privacy more generally.

Addendum: I am aware of concerns in regards to paternity. On the whole I generally think in most situations this is probably information that is going to come out in any case, and so it wouldn’t hurt for it to emerge earlier. Additionally, in the cases of historical figures such as Thomas Jefferson’s presumed line of descent there were widely diverging views among the white descendants as to whether they should cooperate because of the possible moral implications. I suspect most would agree it is better to know this information, even though it implied that line of putative black Jefferson descendants may have paternity misassignment in their lineage. Finally, obviously these issues are far diminished in the case of mtDNA, since maternity is guaranteed. Though one never knows if someone who was adopted was never told of his reality.

300,000 Scoville units

- Da Bomb Ghost Pepper. This is a relatively mild sauce, and is more notable to me for its saltiness than anything else. The label says ~30,000 Scoville units, and that seems about right.

- Da Bomb Beyond Insanity. At 100,000 Scoville units it’s taking it to the next level. I can’t really say there’s much distinctive for me about this sauce. It sufficed in terms of the spice, but it wasn’t exceptional in either that or flavor.

- Da Bomb Ground Zero. Definitely not the least. At 300,000 Scoville units this will test a hardened pepper hound, but, it’s not physically dangerous. The taste isn’t as artificial and metallic to me as the Dave’s Insanity line. This is the closest in terms of utils to fresh habanero.

- Finally, Da Bomb: The Final Answer. I enjoyed this…but this sauce is a serious hazard. To be safe and comfortable you should handle it with disposable gloves, and store it somewhere that the unsuspecting won’t be able to stumble upon. In the end I have to go with Ground Zero because it’s not a pain in the ass. The Final Answer is really just a “stunt sauce’ in terms of usage as a condiment. Of course it would still be useful for cooking, but then I have to ask why you just don’t go with a pure capsaicin extract of some sort.

Credit: Antonio Zulgaldia

But it’s not just genetics. Some people have asserted that Google Glass is a replacement for smartphones. If so, we should be concerned if we’re concerned about privacy more generally. For example a few months ago I was asking a friend of mine the age of her father. She didn’t have any idea with any degree of precision, so I just looked up who she was related to using free online databases with my Android phone. Her father’s age was listed right there. Additionally I was surprised to find that he’d lived in the Pacific Northwest 30 years ago, though my friend seemed uninterested in this biographical data. Rather, she was a touch alarmed that within 30 seconds I retrieved this sort of information, and incidentally noted her sisters’ names, residences and employment histories.

Imagine this sort of functionality, and more so, integrated with something like Google Glass. Yes, I understand that initial privacy concerns will mute the creepier possibilities, but it’s likely a matter of time before someone enables functionality which is initially forbidden. Google won’t have a monopoly on this technology indefinitely. In terms of interpersonal relationships one could easily imagine artificial intelligence which is optimized toward tracking the eye movements of others and constantly outputting a stream of analytics toward the end user. As a concrete example consider a woman who is aware that her significant other has a history of straying. In social situations she could simply turn on a “head tracker” which would generate a frequency distribution over time of who he was looking at over the course of an hour at a party. Of course this sort of thing occurs intuitively and ad hoc already, but with the raw data recorded one might be able to generate much more powerful, persuasive, and incriminating inferences. This might elicit an adaptive response on the part of her boyfriend, but that might be the aim in the first place!

Starting ~10,000 years ago mankind took the step which introduced us to the world of privacy. Rather than small to medium sized bands and villages governed by something like Dunbar’s number we had the option of anonymity. In the 20th century urban life has allowed for a possibility of relatively withdrawal from social contacts and connections, if one so chooses. Such a choice was not available to our ancestors, who were inextricably dependent upon their social network to buffer them from the vicissitudes of fate. The scenario which I’m outlining above, which I think is highly likely, does not correspondent in the details to the ancient villages. Rather, the modern global village expands the scope of those in your potential social network up to whole world.

The reality is of course that you won’t know billions and billions. But in crowded urban societies you’ll have access to personal details and information on nearly everyone you interact with, which may run into tens of thousands per year, which is greater by orders of magnitude from the low hundred posited for our ancestors. We shall adapt, I have no doubt about that, but how is the question. I envisage that some will become “privacy Amish,” creating retro communities bound together by the possibility of anonymity. For the majority there will emerge new rules and norms as to what is gauche and what is polite. Interesting times.

The New York Times

The GSS has two variables which query this issue crossed with ideology rather well. Here’s the query so you can replicate:

Row: NUKEELEC NUKEGEN

Column: POLVIEWS(r:1-3″Liberal”;4″Moderate”;5-7″Conservative”)

Selection filter(s): year(2010-*)

As you can see liberals do tend to be more skeptical of nuclear energy, but it is not stark. In fact, attitudes toward nuclear power seem to be as strongly, if not more so, variant on a populist vs. elite axis than conventional ideology. Here’s the second question replicated for education:

But, when you look only at college educated individuals the ideology divide doesn’t go away. In fact, it seems more extreme.

That’s strong circumstantial evidence that the gap here is one of cultural norms and values, and not facts.

For the record, I don’t think we should intervene in Syria at all. Most Americans seem to agree with me. But it may not matter. The establishment may force intervention.

The artist

It’s been about a year and a half since I officially became a father. I put the official qualifier there because I knew I was going to become a father about two years ago, and many of the psychological changes probably began then. My own reflections and lessons are obviously influenced by my own specific situation. I am not the primary caregiver. It would be too pat to say that our family is the typical college educated sort in all its details, but it is not that far from the truth. My daughter, and her parents, have resources, both financial and familial, which are not there for about half of Americans. I obviously can’t speak to the struggles of working class single mothers. And the American class system being what it is I can’t say I know any such women very well beyond the level of tenuous acquaintanceship.

The first reflection I want to admit is that it is one thing to assert the power of hertability in the abstract, and another to see it in the concrete. It is easy enough to do a scatterplot of a quantitative trait like height. But at very young ages of much more salience are mannerisms and other less quantifiable behavioral tics. It was immediately obvious within the first month of her life that my daughter has some of her parents’ idiosyncratic habits. To give an example there is a way I pick my fingers, in particular my thumb, without even thinking about it. I share this trait with my mother. My daughter began to exhibit the same behavior in her sleep when she was less than one month old. Similarly, there are a variety of body movements where she reminds me a great deal of my youngest brother…but that is perhaps more a function of the fact that he is the one with whom I share the most in terms of biomechanical affect.

Probably one thing that helps keep my inferences and pattern matching in context and in check is the fact that I’ve read The Nurture Assumption. The central observation of this book is that only about ~10% of differences between children on behavioral traits seems to be due to unique parenting strategies and tactics. In other words, your choices as a parent don’t matter nearly as much as you think. It turns out that about half the remaining variation is due to the genes you contribute, and the other half is unaccounted for. To many of you this may seem a trivial insight, but it isn’t. Our media is saturated by arguments at the heart of which is the assumption that the role of parents is central to the development of our children. The current fad for attachment parenting is a case in point. All of these cultural currents hook into the reality that as parents you do agonize often about the most minor details of your child’s upbringing, and what effects that might have in the long term. This makes sense to be, as much of life is a positional game. I have no reason to doubt that this wasn’t true in the past, in which case the marginal impact of parenting on the individual level may have been quite significant. 10% might not seem like so much, but that’s the difference between a B+ and an A+!

Click to enlage

But the major problems with this assumption of the centrality of parental behavior and modeling crop up when you assume they are incredibly causal on a broader social level. This article in The Atlantic is typical, The Distinct, Positive Impact of a Good Dad. Before disputing some of the interpretation of the data I want to admit that all things equal I believe it is better to have two parents than just one. The problem is that not all things are equal. If you click to enlarge the infographic you will see that it states that “An asterisk (*) indicates a statistically-significant difference (p < 0.05) indicated between the group and those who scored in the top third of relationship quality with their father, controlling for respondent’s age, race/ethnicity, level of mother’s education, and household income.” There is one control I do not see listed there: genetics.

A stylized example will illustrate the problem as I see it. Imagine a random sample of 18 year old women who become pregnant by their boyfriend. Half of the boyfriends refuse to help support the child or enter into a lasting relationship, while half of them do so. After 10-20 years you tally the outcomes of the offspring, and you see that those whose fathers remained active in their lives had far better life outcomes when measured on broad metrics of social dysfunction such as delinquency. One inference you could make is that the environment fathers provide is essential toward stabilizing the home and allowing their children to flourish. That is the conventional conclusion that is normally made both in public and private. And, that is why you control for facts such as mother’s education and income in the result above.

But if you think about it closely you see a rather obvious assumption in the model: that the genetic dispositions of the men who abandoned and those who did not abandon are the same. The sort of “man” who would abandon the mother of his child and his child is likely to have a disposition which will be passed down toward his children, even in his absence. The main blind spot with the idea that “family values” is a panacea for all of our problems is that proponents often give nurture all the credit, and don’t even consider the possibility of heritable dispositions. Behavior genetic work tends to point to the possibility that in fact these heritable dispositions strongly effect life outcomes. Fathers matter, but for rather diverse reasons. Neither his sperm nor his magical presence explain it all.

And yet please remember that a huge fraction of the variation is not accounted for. It doesn’t seem to be straightforwardly heritable, nor is it due to distinct home environment. Judith Rich Harris in The Nurture Assumption proposed that much of the “non-shared environment” was peer group. A simple illustration of the dynamic has to do with accents. Children usually speak with the accent of their peer cohort, not that of their parents. But the reality is that we still don’t know what this non-shared environmental component of variance is. In  fact it could be gene-gene interactions, which won’t get captured by measures of heritability. Or it could be developmental stochasticity. I bring these up to point out that even if less than half of the variance of the trait in the population is due to genes that does not entail that it is particularly amenable to reshaping via social policy. Secondarily, even highly heritable traits, such as height (~80% of the trait variation is due to genetic variation) exhibit only mild correlations when you look across siblings (r ~ 0.50 for height). So lesson two: children are going to vary, and you may not have much control over that variation.

Ultimately what my daughter has taught me as a father intellectually in regards to raising her is that I always have to update my assumptions and beliefs, and allow her to give me input into her own development. I don’t subscribe to any major “school” of parenting, nor do I think there is a one-size-fits-all model which would be appropriate for the majority of children beyond the basics (i.e., feed your children adequately and prepare them for the expectations of society as a whole). But natural and social science does give me broad parameters, setting the horizons of the landscape. And yet the space of possible choices is still enormous. For me my daughter is a sort of personal tachyon, her existence is so absorbing that I have a hard time viscerally recalling when she did not exist. And I wouldn’t have it any other way. But that’s feeling, not analysis. And sometimes that is the point!

Credit: Robert Payne

Observe the large variance in ancestry of Diana’s two third cousins presumably derived from Eliza Kewark (though there is always the chance that these segments come from different South Asian ancestors, the typically South Asian mtDNA match across the two reduces the probability of that being the answer in this case). Beyond eight generations the chance of a genetic segment being passed from an ancestor down to a descendant is small. Diana’s cousins are seven generations down from Eliza Kewark, so it isn’t totally implausible that a segment should get passed down. But William at eight is at the boundary, and he may carry no segments (in fact, Diana may have carried no segments). Of course I did note that their mtDNA is likely to be passed down, because there is no element of chance in that. You have your mother’s mtDNA. But one can debate whether mtDNA, which is not present in the nucleus, really counts as ancestry. I believe that heritable genetic material is heritable genetic material. Assuming the lines of descent are as they are recorded I accept that we know for a fact that William likely has South Asian mtDNA. But we most certainly do not know if he has any South Asian autosomal DNA.

But in the end how much does this matter? People will make of it what they will. And yet there is an important aspect to note: this seems like another instance of the firm BritainsDNA hyping genetic findings to increase their profile. You see that the screenshot of their website shows that they’re promoting the story about William’s ancestry, and, they’re also claiming that they are offering the world’s most advanced genetic ancestry test. First, I have to observe that their price points are very high, and second, if they are going to claim the most advanced test in the world they should actually do a point-by-point comparison with other services, which they don’t seem to do.

Perhaps more importantly this outfit now has a history of getting caught up in, and frankly stoking, hype. I don’t begrudge anyone their livelihood, or financial success, but scientists do have an implicit and explicit honor code. Jim Wilson has been involved in some interesting work, which is what I knew him from. But a third time will be a trend, and Wilson shouldn’t be surprised if rather soon he becomes thought of as a ‘tabloid geneticist,’ rather than a scholar who popularizes serious science to the broader public.

Prime Minister of the United Kingdom, 1812 – 1827

“A naturally occurring DNA segment is a product of nature and not patent eligible merely because it has been isolated,” Justice Clarence Thomas wrote for a unanimous court. But manipulating a gene to create something not found in nature is an invention eligible for patent protection.

The case concerned patents held by Myriad Genetics, a Utah company, on genes that correlate with increased risk of hereditary breast and ovarian cancer.

Believe it or not USA Today was out fast with a long story on this issue with quotes and everything. Here’s the full text of the decision. Needless to say this is a pretty big deal, and I’m somewhat surprised this was a unanimous decision. Perhaps the justices actually listen to scientists and their bleating sometimes?

I’ll be checking in to the Genomics Law Report regularly today….

Update: Just to note, several friends have noted that aspects of the science in the ruling seem to have some howlers. That is not surprising (see Scalia’s admission of ignorance in the concurrence). But from listening to the panel discussion on the Myriad case at ASHG 2012 this ruling is still a huge step forward. Being wrong is preferable to “not even wrong.”

MDS of all samples

Click to enlarge

Reading the paper it’s quite obvious that just merging the samples was a big feat. And it’s not just sample size, they had excellent population coverage (267). As Dienekes observes microsats are somewhat “retro”, but try and get this sort of population coverage with whole genomes, or even SNPs. You can get to N>5,000, but with SNPs the overlapping markers start to drop off very quickly, to the point where they are far less informative than this number of microsats. Dienekes quite liked the tree to the left, and I’ve uploaded a rather large version of it for your enjoyment (just zoom in if your browser sizes it down).

But to some extent the tree above illustrates the limitations of this sort of analysis. Rather than an analysis, this is really more a useful data set that you have to slice and dice, and explore on a finer grain. Pooling all the samples together makes it far less informative and unintelligible. This is already obvious in their aggregation to create the large data set, as they had to prune very large subpopulations so they didn’t overwhelm the results. Even then problems obvious to those familiar with the data crop up, though they might not be so clear to those who are reading superficially. The Gujarati data set among the South Asians separated out on a two dimensional visualization from all other populations. This is something that often occurs because it looks like Gujaratis are sampled from a very specific caste, which increases the perceived affinity of this regional ethnicity. Similarly, pooling all the populations and representing them on a two dimensional plot is more an aesthetic declaration than an informative visualization. You have to bracket out the populations to see value-added structure. Finally, even the coarse and general observations need to be integrated with caution. Rosenberg’s lab has been illustrating the decay of genetic diversity from Ethiopia for nearly a decade now. It’s a classic result which shows up in graduate level population genetics courses. But both the anthropology and genetics tell us that Ethiopians are a compound population with Sub-Saharan African and Eurasian affinities. Most readers can be expected to know this, but I would not be surprised if some simply took the general plot at face value and applied the insight to all the populations, as if they really were subject to a serial founder effect (my specific point is that Ethiopians are the product of a synthesize due to back migration, reversal of the general migration out of Africa being illustrated with the decline in genetic diversity).

Overall I find this an interesting paper which sets the backdrop for understanding the canvas of human genetic variation. The only last caution I would offer is that microsatellites are atypical regions of the genome which evolve rapidly in a neutral fashion. This makes them excellent for pinpointing population differences and inferring history from a limited marker set. But I think people should be cautious of specific novel results, and not hold them up as that authoritative when we have high density SNP data.

Note: They’ve released the data. If readers are curious about doing different things with these data than was shown in this paper, Treemix can handle microsats. Also, props to them for releasing this creative commons.

Citation: Pemberton, Trevor J., Michael DeGiorgio, and Noah A. Rosenberg. “Population structure in a comprehensive genomic data set on human microsatellite variation.” G3: Genes| Genomes| Genetics 3.5 (2013): 891-907.

While I’m at it, I notice there are some affordable copies ($10-$20) of Genetics of Human Populations. This is a very old work from decades ago. But it is an encyclopedic treatment of the post-World War II human genetic literature, much of which has been forgotten or faded, but perhaps should not be discarded so lightly….

Update: I realize on second thought that it was remiss of me to not point out that Graham Coop and Joe Felsenstein both have free population genetics resources which readers might find highly useful. Also see these notes from U Conn.

Sir Francis Galton

Modern evolutionary genetics owes its origins to a series of intellectual debates around the turn of the 20th century. Much of this is outlined in Will Provines’ The Origins of Theoretical Population Genetics, though a biography of Francis Galton will do just as well. In short what happened is that during this period there were conflicts between the heirs of Charles Darwin as to the nature of inheritance (an issue Darwin left muddled from what I can tell). On the one side you had a young coterie around William Bateson, the champion of Gregor Mendel’s ideas about discrete and particulate inheritance via the abstraction of genes. Arrayed against them were the acolytes of Charles Darwin’s cousin Francis Galton, led by the mathematician Karl Pearson, and the biologist Walter Weldon. This school of “biometricians” focused on continuous characteristics and Darwinian gradualism, and are arguably the forerunners of quantitative genetics. There is some irony in their espousal of a “Galtonian” view, because Galton was himself not without sympathy for a discrete model of inheritance!

William Bateson

In the end science and truth won out. Young scholars trained in the biometric tradition repeatedly defected to the Mendelian camp (e.g. Charles Davenport). Eventually, R. A. Fisher, one of the founders of modern statistics and evolutionary biology, merged both traditions in his seminal paper The Correlation between Relatives on the Supposition of Mendelian Inheritance. The intuition for why Mendelism does not undermine classical Darwinian theory is simple (granted, some of the original Mendelians did seem to believe that it was a violation!). Many discrete genes of moderate to small effect upon a trait can produce a continuous distribution via the central limit theorem. In fact classical genetic methods often had difficulty perceiving traits with more than half dozen significant loci as anything but quantitative and continuous (consider pigmentation, which we know through genomic methods to vary across populations mostly due to half a dozen segregating genes or so).

Credit: Richard Wheeler

Notice here I have not said a word about DNA. That is because 40 years before the understanding that DNA was the substrate of genetic inheritance scientists had a good grasp of the nature of inheritance through Mendelian processes. The gene is fundamentally an abstract unit, an analytic element subject to manipulation which allows us to intelligibly trace and predict patterns of variation across the generations. It so happens that the gene is instantiated in a material sense through sequences of the biomolecule DNA. This is very important. Because we know the material basis of modern genetics it is a much more fundamental science than economics (economics remains mired in its “biometric age!”).

The “post-genomic era” is predicated on industrial scale analysis of the material basis of genetics in the form of DNA sequence and structure. But we shouldn’t confuse DNA, concrete bases, with classical Mendelism. A focus on the material and concrete is not limited to genetics. In the mid-2000s there was a fad for cognitive neuroscience fMRI studies, which were perceived to be more scientific and convincing than classical cognitive scientific understandings of “how the mind works.” In the wake of the recession of fMRI “science” due to serious methodological problems we’re left to fall back on less sexy psychological abstractions, which may not be as simply reduced to material comprehension, but which have the redeeming quality of being informative nonetheless.

This brings me to the recent paper on SNPs associated with education in a massive cohort, GWAS of 126,559 Individuals Identifies Genetic Variants Associated with Educational Attainment. You should also read the accompanying FAQ. The bottom line is that the authors have convincingly identified three SNPs to explain 0.02% of the variation in educational attainment across their massive data set. Pooling all of the SNPs with some association they get ~2% of the variation explained. This is not particularly surprising. A few years back one of the authors on this paper wrote Most Reported Genetic Associations with General Intelligence Are Probably False Positives. Those with longer memories in human genetics warned me of this issue in the early 2000s. More statistically savvy friends began to warn me in 2007. At that point I began to caution people who assumed that genomics would reveal the variants which are responsible for normal variation on intelligence, because it seemed likely that we might have to wait a lot longer than I had anticipated. As suggested in the paper above previous work strongly implied that the genetic architecture of intelligence is one where the variation on the trait in the normal range is controlled by innumerable alleles of small effect segregating in the population. Otherwise classical genetic techniques may have been able to detect the number of loci with more surety. If you read Genetics of Human Populations you will note that using classical crossing techniques and pedigrees geneticists did in fact converge upon approximately the right number of loci segregating to explain the variation between European and African pigmentation 60 years ago!

Some of my friends have been arguing that the small effect sizes here validate the position that intelligence variation is mostly a function of environment. This is a complicated issue, and first I want to constrain the discussion to developed Western nations. It is an ironic aspect that arguably intelligence is most heritable among the most privileged. By heritable I mean the component of variation of the trait controlled by genes. When you remove environmental variation (i.e. deprivation) you are left with genetic variation. Within families there is a great deal of I.Q. difference across siblings. The correlation is about 0.5. Not bad, but not that high. Of course some of you may think that I’m going to talk about twin studies now. Not at all! Though contrary to what science journalists who seem to enjoy engaging in malpractice like Brian Palmer of Slate seem to think classical techniques have been to a great extent validated by genomics, it is by looking at unrelated individuals that some of the most persuasive evidence for the heritability of intelligence has been established. It is no coincidence that one of the major authors of the above study also is an author on the previous link. There is no contradiction in acknowledging difficulties of assessing the concrete material loci of a trait’s variation even if one can confidently infer that association. There was genetics before DNA. And there is heritability even without specific SNPs.

Additionally, I want to add one caveat into the “environmental” component of variation. For technical reasons this environmental component may actually include relatively fixed biological variables. Gene-gene interactions, or developmental stochasticity come to mind. Though these are difficult or impossible to predict from parent to offspring correlations they are not as simple as removing lead from the environment of deprived children. My own suspicion is that the large variation in intelligence across full siblings tell us a lot about the difficult to control and channel nature of “environmental” variation.

Finally, I want to point out that even small effect loci are not trivial. The authors mention this in their FAQ, but I want to be more clear, Small genetic effects do not preclude drug development:

Consider a trait like, say, cholesterol levels. Massive genome-wide association studies have been performed on this trait, identifying a large number of loci of small effect. One of these loci is HMGCR, coding for HMG-CoA reductase, an important molecule in cholesterol synthesis. The allele identified increases cholesterol levels by 0.1 standard deviations, meaning a genetic test would have essentially no ability to predict cholesterol levels. By the logic of the Newsweek piece, any drug targeted at HMGCR would have no chance of becoming a blockbuster.

Any doctor knows where I’m going with this: one of the best-selling groups of drugs in the world currently are statins, which inhibit the activity of (the gene product of) HMGCR. Of course, statins have already been invented, so this is something of a cherry-picked example, but my guess is that there are tens of additional examples like this waiting to be discovered in the wealth of genome-wide association study data. Figuring out which GWAS hits are promising drug targets will take time, effort, and a good deal of luck; in my opinion, this is the major lesson from Decode (which is not all that surprising a lesson)–drug development is really hard

Addendum: Most of my friends, who have undergraduate backgrounds in biology, and have taken at some quantitative genetics, seem to guess the heritability of I.Q. to be 0.0 to 0.20. This is just way too low. But is it even important to know this? I happen to think an accurate picture of genetic inheritance is probably useful when assessing prospects of mates….

Citation: Rietveld, Cornelius A., et al. “GWAS of 126,559 Individuals Identifies Genetic Variants Associated with Educational Attainment.” Science (New York, NY) (2013).

This isn’t rocket science, so to speak. Information dissemination is pretty easy right now, and that is theoretically one of the major things which drives science. This should be a great time for scientific progress! Is it? In genomics, yes, though that’s not because of more efficient flow of information, as opposed to technology. With that prefatory comment, I think John Hawks’ recent jeremiad is worth reading, Speak up and matter:

I have little patience for the risk-averse culture of academics.

The bottom line is: People need to decide if they want to be heard, or if they want to be validated.I have long been an associate editor at PLoS ONE, and once I edited a paper that received a lot of critical commentary. That journal has a policy of open comment threads on papers, so I told disgruntled scientists to please write comments. The comments appear right with the article when anybody reads it, they appear immediately without any delay, and they can form a coherent exchange of views with authors of the article and other skeptical readers.

Some of the scientists didn’t want to submit comments, they wanted to have formal letters brought through the editorial review process. “Why?” I wrote, when you could have your comments up immediately and read byanyone who is reading the research in the first place? If you want to make an impact, I wrote, you should put your ideas up there right now.

They replied, “How would you feel if someone published something wrong about Neandertals? Wouldn’t you want to publish a formal reply?”

I wrote: “In that case, I would probably get a blog.”

What is the difference between being heard and being validated? It’s whether you are contributing to the solution or to the hindsight.

Academics are risk averse. I’m thinking here of a friend who expressed a bit of paranoia about maintaining good relations with his Ph.D. adviser because negative comments from that direction can still matter ten years down the line in tenure reviews. Science is highly political, and money and career options are finite. You have to be careful about who you might offend.

Or do you? What’s the point of doing science if security is what you want? I think scientists need to be careful about personalizing a war of ideas into a war of people, though this is going to happen because science is a human endeavor. But there are many easier jobs than science which pay better. Those jobs require political skills and machinations, but many of them don’t put forward the pretense of a noble elevation of the truth above all.

One day we all die. On that day if you are a person interested in forwarding science and understanding the world, being true to truth, do you want too many regrets about how you had to “work the system” and cultivate relationships? If that was an overwhelming focus rather than speaking plainly, then you should have gone into a more lucrative career. The world needs more actuaries.

Addendum: Much of the same could be said about journalism. And science journalism? Well….

Credit: Puma

For replication here are the variables:

Row: EATGM POLINFGM BIZINFGM MEDAGRGM MEDINFGM GMMED GMPOL GMBIZ POLINFNK

Column: POLVIEWS(r:1-3″Liberal”;4″Moderate”;5-7″Conservative”)

There results are presented below (rows add up to 100% for each question).

What this tells us is that elite opinions matter a lot in public discourse. The gap between liberals and non-liberals is not really there on this issue at the grassroots. That could change, as people of various ideologies tend to follow elite cues. This is why the strong counter-attack from within the Left elite is probably going to be effective, as it signals that being against GMO is not the “liberal position.”

Addendum: Just so people who haven’t been reading me know, political moderates tend not to be very intelligent.

Anyone who reads the genomic posts with any interest on his weblog must read Daniel Lawson’s fine review of the topic which he has posted on arXiv, Populations in statistical genetic modelling and inference (via Haldane’s Sieve). Even if you don’t have a population genetic and genomic background the gist is entirely accessible. If you do have a population genetic and genomic background and haven’t used various packages such as STRUCTURE or EIGENSOFT yourself, I would recommend reading Lawson’s characterizations, as they are all spot on.

Also, if you have not, I recommend Lawson’s website for ChromoPainter and fineSTRUCTURE. The utility of these methods is outlined in the paper Inference of Population Structure using Dense Haplotype Data.

Fetus at ~12 weeks

I am going to get back to the eugenics debate at some point, but it is hard to motivate myself. This is due to a combination of complacency and sanguinity. Many of those who use eugenics as a “scare word” or are “very concerned about it” don’t really seem to get past generalities when it comes to the present situation (i.e., there is detailed exploration of past atrocities, and some exploration of rather unrealistic scenarios, such as occurred with the “Chinese eugenic” story, but little concrete engagement with realities such as the high abortion rates for positive tests for Down syndrome). In more crass and intellectually vapid discussions liberals and conservatives tend to use eugenics as a term of selectively useful instrumental rhetoric, a bludgeoning instrument only in the mindless screaming discourse.

Meanwhile, we have advances like the whole genome sequencing of second trimester fetuses. This is still basic science, but in genomics basic science is translated really fast to the consumer market. I’m ~90 percent sure my daughter will have a 10 x whole genome sequence by the end of 2014 (I might even get her parents in on the game for a trio). So, submitted for your interest are two papers on first trimester noninvasive screens for Down syndrome due to aneuploidies (and other syndromes). Non-Invasive First Trimester Blood Test Reliably Detects Down’s Syndrome and Other Genetic Fetal Abnormalities:

An Ultrasound in Obstetrics & Gynecology study by Kypros Nicolaides, MD, of the Harris Birthright Research Centre for Fetal Medicine at King’s College London in England, and his colleagues is the first to prospectively demonstrate the feasibility of routine screening for trisomies 21, 18, and 13 by cfDNA testing. Testing done in 1005 pregnancies at 10 weeks had a lower false positive rate and higher sensitivity for fetal trisomy than the combined test done at 12 weeks. Both cfDNA and combined testing detected all trisomies, but the estimated false-positive rates were 0.1% and 3.4%, respectively.

…

A second Ultrasound in Obstetrics & Gynecology study by the group, which included pregnancies undergoing screening at three UK hospitals between March 2006 and May 2012, found that effective first-trimester screening for Down’s syndrome could be achieved by cfDNA testing contingent on the results of the combined test done at 11 to 13 weeks. The strategy detected 98% of cases, and invasive testing was needed for confirmation in less than 0.5% of cases.

We did CVS. We’d rather not have to in the future. The key is to move positive tests into the first trimester. No matter the reality that most couples who receive a positive result in early second trimester choose to terminate, when it comes to killing a fetus every week counts. You can see what I mean when you look at how abortions are performed as a function of fetal development. Not only that, if you read the source papers you see that the typical woman who receives these sorts of screens is 35-40 years of age, and in that case with the fertility clock ticking every week is of the essence.

Citations:

1. Implementation of maternal blood cell-free DNA testing in early screening for aneuploidies.Ultrasound in Obstetrics & Gynecology, 2013; DOI:10.1002/uog.12504
2. First-trimester contingent screening for trisomy 21 by biomarkers and maternal blood cell-free DNA testing. Ultrasound in Obstetrics & Gynecology, 2013; DOI:10.1002/uog.12511

Christopher Columbus

A few year ago there was a minor controversy when some evolutionary genomicists reported that they had reconstructed the genome of the extinct Taino people of Puerto Rico by reassembling fragments preserved in contemporary populations long since admixed. The controversy had to do with the fact that some individuals today claim to be Taino, and therefore, they were not an extinct population. Though that controversy eventually blew over, the methods lived on, and continue to be used. Now some of the same people who brought you that have come out with work which reconstructs the recent demographic history of the Caribbean, both maritime and mainland, using genomics.  Even better, it’s totally open access because it’s up on arXiv, Reconstructing the Population Genetic History of the Caribbean (please see the comments at Haldane’s Sieve as well, kicked off by little old me). Though the authors pooled a variety of data sets (e.g., HapMap, POPRES, HGDP) the focus is on the populations highlighted in the map above.

Much of the novel insight in the results begins with their observation of a distinct “Latino” population genetic cluster with strong affinities with Europe within the Caribbean populations. This is clearly visible in their ADMIXTURE analysis. What they did was pool various populations, and run a method which decomposes the ancestry of each individual as a combination of K ancestral populations. In cases where the pooled populations are clear and distinct the results will be clear and distinct. For example, if you had 50 Finns and 50 Nigerians and pooled them, and ran ADMIXTURE at K = 2, then with a non-trivial number of SNPs (10,000 is more than sufficient) all the Finns and Nigerians will partition into two distinct ancestral populations according to these sorts of model based clustering. But it always has to be remembered that though these methods map onto reality, and give us some sense of the variation within the data sets, the K’s themselves are artificial constructs. So, for example, the HGDP Maya population is known to have non-trivial European gene flow. If you use this sort of Maya population as your “Native American” reference, then you will underestimate Native ancestry in admixed groups because your reference Native population is already skewed toward Europeans (this is obviously a major problem when you don’t have the appropriate reference because it is extinct, such as with the Taino).

With those cautionary preliminaries out of the way what’s going on in these results? As you can see many of the Caribbean populations are straightforward combinations of various continental ‘parent’ populations. This is clearly evident in K = 3, where green = Africa, red = European, and blue = Native (note that the Maya have a range of European ancestry just as I said). By looking at individual variation within populations you can already gain some insights as to the nature of the admixture. In Mexico there is a wide range of the European vs. Native fraction, though in this data set there are no “pure” individuals. Additionally, there are low, but relatively even, amounts of African ancestry across the population. Though African consciousness this is not a major element of modern Mexican national identity, people of African ancestry were a major part of the Spanish colonial enterprise (see Empire: How Spain Became a World Power, 1492-1763). In some areas, such as Veracruz, people of visibly African ancestry remain, but in much of Mexico these individuals intermarried and their physical characteristics were diluted toward the point of not being visible.

The situation in the maritime Caribbean is somewhat more complex. In these contexts it was the Native, not African, ancestry which was subsumed and submerged. It is genomics which has ‘rediscovered’ this ancestry, to the extent that many scholars had previously been skeptical of the possibility that modern Puerto Ricans and Dominicans inherited a substantial share of Taino ancestry. In both Puerto Rico and the Dominican Republic the relevant issue is that there is a wide range of proportion of African and European ancestry, with Cuba being the notable extreme case of this phenomenon. What’s going on with Cuba in particular is that there were late waves of migration from Spain, so some modern white Cubans are much less affected by admixture than other Caribbeans (remember that Cuba was part of Spain until 1898). In Haiti the situation is reversed, where the revolutions of the late 18th and early 19th centuries had a racial tinge, and whites were expelled (leaving a small mulatto class).

But it is K = 8 where things really get interesting. The black component is a European Iberian-like element which is distinct to Latino populations (including Maya). As you can see on this PCA the Latino element is related to the Iberian populations, as they took the European segments from the Caribbean populations and used them to flesh out the distribution in ancestry. There are several ways to interpret this. Dienekes suggested this might simply be a function of the source Iberian populations hundreds of years ago being somewhat different from the contemporary ones. For example, obviously contemporary Spaniards would be more subject to gene flow with other Europeans >1600 than their New World cousins. Another possibility is that there was extreme sampling from a particular region of Spain, and that has how broken out as its own cluster. For example, I know that a disproportionate number of migrants were from Andalucia and Extramadura. But the pattern here doesn’t suggest to me that possibility (the black dots should be more south-shifted I would think if they were from those two provinces).

Rather, the interpretation they seem to favor is that this element has been drifted away from the ancestral populations due to a bottleneck. This is not ethnographically implausible; the early years of the Spanish colonial experiment was characterized by de facto polygyny. Many adventurers lived lives not unlike those of the white grandees of the East India company in the late 18th century. Some have argued that this period of ubiquitous common law polygyny has influenced the fact that illegitimate births have traditionally been very common in Latin America. One reason the authors favor the bottleneck model is that the genetic distance between the Latino element and the Iberian one is rather high. This is often common in situations where drift/bottleneck has deviated allele frequencies particularly rapidly. Not only that, but the tendency is most strong in maritime Latin America, many of whose islands received relatively fewer subsequent migrants than the large and expansive mainland viceroyalties.

23andMe ancestry decomposition for friend who is 1/4 Asian

Another way the authors explored the demographic history was to look at the length distribution of the tracts of ancestries. How this works is simple. A first generation hybrid will have unbroken lengths of ancestry each parent, but subsequent generations will start to have fragmentation occur as recombination breaks apart long blocks identical by descent. You can see this in the figure to the left, where my friend who has one Asian grandparent has blocks of alternating European and Asian ancestry because of meiotic recombination events. The longer from the time of admixture the smaller and smaller the blocks will become, as recombination slices apart long blocks and recombines ancestral components. By looking at the distribution and mix of lengths the authors can construct demographic histories of the populations. In short it looks like much of the European ancestry came in one short quick pulse, rather early on in settlement. This is in keeping with the high reproductive output attested for European males thanks to polygyny during this period.

The same method was performed for the African ancestry, and the authors discovered an intriguing result. It seems that in the early years most of the Caribbean black slaves were derived from the western tip of Sub-Saharan Africa, from the Senegal river down to modern Ghana. Later on the longer tracts show affinities with populations further east, from the Bight of Benin toward the Equator. I don’t know the history of slavery well enough to confirm or deny the reality of this finding, but it illustrates the power of genomics combined with wide sampling strategies. More relevantly I suspect genomics’ role will be to assign magnitudes to known dynamics.

Finally, the authors also inferred diverse relationships for the Native admixture in the Caribbean populations. They confirmed some evidence of south-to-north migration into Central and Caribbean America, and also specific ethno-linguistic associations between now de facto extinct Caribbean populations and those of mainland South America. Some of these results have long been suggested, but lack of historical documentation makes inferences shadowy. Genomics can not resolve these debates, but they shed light upon them.

Overall this is an interesting study because I think it is a test run at the sort of historical-demographic questions that genomics will be used for. There has long been a ‘genetics as a tool’ school of thought among many ecologists and phylogeneticists, and now you shall have a ‘genomics as a tool’ to sit right along side that in many more diverse fields. Caribbean and Latin American populations are the low hanging fruit, because the Spanish and Portuguese colonial experiment are reasonably well attested, and the source populations are very distinct (so easy to pick signal out of the noise). But there are other historical questions of the same period which are also of interest. In Albion’s Seed David Hackett Fisher describes four Anglo-American folkways which contributed to the culture of this nation. Of these, ~20,000 Puritans arrived between 1620-1640 and became the ancestors of ~700,000 by 1970. Though 20,000 is not quite a bottleneck (in fact, they arrived from different sectors of England), I am curious if these individuals, a segment of “Old Americans,” can still be discerned in the genomic data. This is just one of many possible questions which will be with reach of answer in the near future….

Credit: PartnerHund

It’s an exciting time for those interested in the evolutionary genomics of the dog. In 2010 a big SNP-array paper came out, Genome-wide SNP and haplotype analyses reveal a rich history underlying dog domestication. Today we’re going whole genome, which is important because many of the SNP-arrays are ascertained on domestic dogs (i.e., they are designed to pick up dog variation, and so may distort our perception of the variation in wolves). Recently I talked about an analysis of the evolutionary genomics of the dog, The genomics of selection in dogs and the parallel evolution between dogs and humans. The main interesting result of that group was to push the divergence of the dog and wolf lineages further back in time, ~30,000 years, in line with some archaeological and mtDNA finds. I did not find their arguments for the origin of the dog in East Asian convincing. Now a new preprint on arXiv, Genome Sequencing Highlights Genes Under Selection and the Dynamic Early History of Dogs, pushes this even further.

First, after reading the paper I recommend the comments at Haldane’s Sieve, where the authors engage in some back and forth. Second, one of the authors put the supplementary information on a Dropbox, so you can get that too. I highly recommend this in particular, because it has detailed methods, code, and also concise but useful explanation of concepts such as D-statistics. Overall the paper breaks down into two broad themes, the phylogenetics and analysis of adaptation and selection. There was many X coverage of an Israeli, Croatian, and Chinese wolf, and a Basenji, Boxer, and Dingo. For the primary analysis the sample sizes were N = 1, but that is not a major issue as they had extremely accurate and precise estimates as to the polymorphism across these individuals because of the repeated coerage.

A major takeaway in terms of demographics for this paper is that it’s complicated. The authors inferred that domestic dogs went through a population bottleneck in the past on the order of one to two magnitudes. Second, wolves also went through a population bottleneck, albeit milder. On the first read through I was surprised by the second finding, but after talking to a canine geneticist I was told that this wolf bottleneck had long been known. The genomics confirmed prior expectations rather than smoking out novel inferences.

Perhaps a more surprising finding is that the ancestor of dogs, and yes, there was one ancestral population, not many, derives from an extinct wolf lineage. Their inference was derived from the fact that the three wolves, sampled from putative regions of dog domestication, all exhibit equal genetic distance from the dogs. Previous work had suggested that dogs may have derived from Near Eastern wolves, while other researchers argued for an East Asian origin. The results here support the proposition that these suggestions are misinterpreting genuine gene flow between local populations of wolves and dogs. The authors detected gene flow between West Eurasian wolves and the western dogs (Basenji and Boxer) and East Eurasian wolves and eastern dogs (the Dingo).

On a minor note, these results also confirm a pre-agricultural origin for the dog, with a divergence of ~11-16 thousand years B.P. across the 95% confidence interval. This is at some discrepancy with the Chinese group, but this may just be an artifact of a different mutation rate parameter. The take home either way is that dogs pre-date agriculture.

But that doesn’t mean agriculture is irrelevant. As far as the adaptation goes there’s a lot here, and I’m not sure that this paper has anything revolutionary in that dimension. First, they confirm that just as in humans there is variation among canids in terms of copy number in the amylase gene conditional on lifestyle. Dingos and Alaskan huskies have very few copies, while ancient West Asian dogs have many. Also, the authors find normal variation in wolves for this trait, implying that amylase polymorphism is part of standing genetic variations.

I will leave it to you to survey the veritable alphabet soup of genes which have been buffeted by natural selection by evolutionary process when it comes to dogs. I’m more curious about variation within dog at this point, as there should be heritable variation there too.

Cite: arXiv:1305.7390v2 [q-bio.GN]

I’m somewhat interested in genetics and evolution, to engage in a bit of understatement. My friends know this, so whenever a genetics and evolution themed story or meme explodes in the media they ask me about it. A large fraction of the time I get irritated, because the media often grasps upon very sensational nuggets, distorts them out of shape, and makes genuine understanding difficult. A few weeks ago it was the story of an artist being able to reconstruct portraits from DNA, credulously reported by NPR and The Smithsonian. As someone who tries to keep up on the latest genetic research in forensic genetics I knew the media depictions of what this individual was doing were simply not realistic. Either the artist in question was a fraud, or the media was engaging in conscious or unconscious misrepresentation and conflation. If Matthew Herper’s reporting is correct, and I see no reason to doubt it, seems more likely the latter than the former. Before that there was the genius Chinese babies meme, the robustness of which is attested to by its interjection into the Geoffrey Miller saga (an update was offered, but it is still notable that the original sensationalism has had more legs than subsequent corrections of that sensationalism). Finally, today there emerged a bizarre critique of weblogs over at Current Biology, which was nicely satirized by Christie Wilcox. It always strikes me as rich when institutions which still publish in print and have reasonable overhead costs (e.g. editors) make a big show of their oversight, but due to their power and prominence they are often invariably the exact sort of organization which is perfectly placed to launch a ridiculous meme in the first place!

So in this vein today Forbes publishes a piece titled How The Human Face Might Look In 100,000 Years. I saw this on Twitter and Facebook, and my first thought after seeing the headline was 1) probably totally unfounded in substance 2) multiple people are going to try and sound me out on this. That’s exactly what happened. When Herper asked on Twitter for thoughts on the piece I responded with a vulgarity and a query as to why the faces of the future had the hair of men and women of 2013. I also suggested that these are more likely to be the faces of the future. Why is the piece so ridiculous? Just click through to the article and you’ll see that the faces of the future look something like Rick Hunter and Lynn Minmay.

As far as the text of the piece it reminds me of the incoherence of mediocre space opera. On the one hand humans have access to all sorts of advanced genetic engineering, and yet they value being human-looking due to a mysterious implicit Orange Catholic Bible. Actually taking the inferences seriously is even more humorous:

Eyes will meanwhile get larger, as attempts to colonize Earth’s solar system and beyond see people living in the dimmer environments of colonies further away from the Sun than Earth. Similarly, skin will become more pigmented to lessen the damage from harmful UV radiation outside of the Earth’s protective ozone.

Does anyone believe that future Titanians will deal with the distance from the sun simply via enlarging their eyes? Even today we have infrared goggles, so one presumes that more modular technologies which one might discard and rotate between are going to be more important. Similarly, the nature of solar radiation is such that one will need more than darker skin. Of course there is likely going to be terraforming, but in that case humans could probably create their own artificial ozone layers as well. For more entertainment read the whole thing.

The bigger issue is that these sorts of projections into the far future are folly. It presumes that while technological advancement accelerates by leaps and bounds, our culture is as static as the Oldowan. I am very skeptical even of 100 year projections! And you should be too. If technological civilization still exists 100,000 years from now, and our species has not gone through cycles of booms and busts, then I presume that these far future individuals will live in worlds far less recognizable than that of Alvin in The City and the Stars. If the world of 98,000 years A.D. is recognizable, then something has seriously gone wrong with technological civilization and human cultural evolution.