Our ancestry as a braided estuary

Ancient DNA figure
Citation: Ancient human genomes suggest three ancestral populations for present-day Europeans

Purple text included by author of this post
Citation: Statistical Methods for Evolutionary Trees.
Citation: Statistical Methods for Evolutionary Trees.

At some point you have no doubt encountered trees of the sort you see to the left. They are incredibly useful visualizations of historical relationships between lineages. Breeding populations. The metaphor of the tree of life was co-opted almost immediately by evolutionary science in the 19th century. On the orders of tens of millions to billions of years the idea of diverging and bifurcating lineages is accurate to a great extent in terms of depicting the dynamics of natural history. But even on this scale the tree masks facts which are not of trivial importance. Horizontal gene transfer means that even very sharply delineated branches of the tree of life may share commonalities across wide regions of the genome. The smaller the value which defines the last common ancestors of two putative lineages, the muddier the image reflected through the lens of the tree becomes. And yet the tree visual metaphor persists when comparing populations which are rather close genetically in an evolutionary sense because of its plain utility. Trees are thick in L. L. Cavalli-Sforza’s History and Geography of Human Genes, which paints the broad and rich landscape of human populations only diverged over the past few tens of thousands of years, our own species.

This is not to ignore the self-evident fact that tips of the branches can eventually converge. Geneticists have long acknowledged, and leveraged, recent admixture between populations long separated by time and space. No one denies that African Americans coalesced out of the relations of black slaves and white settlers. Or that the population genetic landscape of Latin America can not be understood without taking into account the varied quanta of African, European, and Amerindian ancestry which defines particular locales. The reality of admixture in these cases was attested to historically, is visible in a straightforward phenotypic sense, and, can be detected using a small number of classical markers.

Read More

Genetic determinism and probabilism

170px-DNA_orbit_animated_static_thumbThe New York Times has a piece up, I Had My DNA Picture Taken, With Varying Results, which begins:

I like to plan ahead; that much I knew about myself before I plunged into exploring my genetic code. I’m a healthy 28-year-old woman, but some nasty diseases run in my family: coronary heart disease, rheumatoid arthritis, Alzheimer’s and breast cancer.

So I decided to read the tea leaves of my DNA. I reasoned that it was worth learning painful information if it might help me avert future illness.

The author is a masters student in bioethics at Columbia (of curious note, her father is the designated heir of Ayn Rand and high priest of mainline Objectivism), so not the typical person off the street. That being said, the piece does reflect implicitly the almost oracular powers conferred to genetic data in the imaginings of the public. It is totally understandable, if somewhat concerning, that direct-to-consumer firms differ in their assessments. Even if the author had obtained a very high quality whole genome sequence, and had a large population sample to compare it against in terms of traits and genotypes, there’s only so much predictiveness you can squeeze out of complex traits. This does not mean that understanding the genetic underpinnings of diseases is without utility. Simply that there are limits to the confidence of predictions of common and complex diseases in the case of one particular individual. In fact, even if you are healthy, have no family history, and exercise, you can die of a heart attack in your 50s.

Not separate, but not equal

A new paper in Nature fleshes out some details about the relationships between Denisovans, Neanderthals, and modern humans, as well as possible others. I believe the figure above gives the flavor of the general findings in terms of phylogenetics, though if you want more I recommend Carl Zimmer in The New York Times. It has to be noted that it’s incredible that such high resolution sequencing could be done on these ancient fossils, when ~10 years ago we were excited about one modern genome. Also, I am struck by, though not surprised, that it doesn’t seem to be that modern humans (our lineage) had too many distinctive major genetic differences from our cousins. Finally, it does seem that the human phylogeny is more properly defined as a graph than a tree. But don’t forget that it doesn’t seem like all the edges were weighted the same. We’re a whole lot more Neo-African than we are Neanderthal or Denisovan.

The Empire of R1a1a and the Levites


R1a1a is one of the most geographically expansive Y chromosomal haplogroups. It spans the Irish Sea to the Bay of Bengal. I am of this lineage, as is my friend Daniel MacArthur. But with deeper exploration of the phylogeny of this haplogroup it seems clear now that it is very diverse, with a great deal of geographic structure. There are a wide range of South Asian lineages, but also one very dominant one in Eastern Europe.

A new paper in Nature Communications, Phylogenetic applications of whole Y-chromosome sequences and the Near Eastern origin of Ashkenazi Levites, addresses a peculiarity in the domain of Jewish genetics. The Levites, the helpers of the Cohen priestly class, seem to be carriers of R1a1a, and this lineage has exploded rapidly in this population (the classic “star-like phylogeny”). The historical genetic question though is this: are the Levites descendents of a Slav proselyte? Within Europe R1a1a exhibits the highest frequency in what was once the Pale of Settlement, so this is a reasonable question.

panelAUsing whole genome analyses and more extensive geographic coverage, the answer to this question seems to be no. Rather, the Levites descend from a distinct West Asian branch of non-European R1a1a. This is evident in panel A of Figure 1. You can see clear that the Ukrainian samples are the outgroup in relation to the other branches of R1a1a. And within those there is further structure, as the South Asian Gujaratis are distinct from the clade in which most of the Ashkenazi Levites are nested (the authors posit that the presence of the Iberians may be attributable to the Moors, plausible enough). The Golden Age of Y chromosomal phylogenies is over, but these markers still have some juice which can be squeezed out.

Citation: Phylogenetic applications of whole Y-chromosome sequences and the Near Eastern origin of Ashkenazi Levites (open access).

Cats: they would be assimilated

Asian wild cat
Asian wild cat

There is a paper in PNAS, Earliest evidence for commensal processes of cat domestication, which raises some interesting questions about the emergence of Felis silvestris catus, the domestic cat. I state questions, because the answers it provides are quite narrow, if plausible. In sum the authors found that ~5,000 years ago a group of cats, possibly Asian wildcats, Felis silvestris ornata, seem to have entered into coexistent relationship with humans. This itself is not entirely surprising. ~10,000 years ago a specimen of Felis silvestris lybica, a Near Eastern wildcat, seems to have been interred at a human grave site on Cyprus. The PNAS paper correctly notes that mitochondrial data also point to Near Easter wildcats has being the wild silvestris group that gave rise to the domestics. Strangely they leave out that autosomal microsatellites, highly variable repetitive regions in the genome, also indicate a clustering of domestic lineages with Near Eastern wildcats (to be clear, this is in the same paper that they cite for the mitochondrial results).

A confluence of results, phylogenetic, phylogeographical, archaeological and morphological, do seem to support the proposition that catus and lybica form a monophyletic clade (or, perhaps more precisely catus is a derived branch of lybica). I won’t put 100% confidence on this, but that’s our best guess. So where do these Chinese results fit into the big picture? You can imagine how the media have reacted. Here’s the marginally staid (by British standards) The Guardian: Ancient Chinese cat bones shake up domestication theory.

Actually there’s that much domestication theory to shake up. The term “domestication” is genuinely problematic in a precise sense, in that it isn’t a precise term. There is a live debate as to whether domestic cats are genuine domestics, whatever that means. Setting this non-trivial semantic aspect of the debate aside, to my knowledge there is no strong genetic evidence that East Asian domestic cats exhibit any signs of deep ancient population structure. That is, that they derive from hybridizations of western and eastern silvestris lineages. That doesn’t mean that it can’t happen. The popular Bengal breed was developed in the past few decades by hybridization domestic cats with Asian leopard cats, which aren’t even in the genus Felis (though the ones adopted out are usually F4 individuals and the product of backcrosses with domestics). But we don’t have widespread evidence of hybridization, though the geographical coverage and marker density leave something to be desired (yes, I’m hedging here, but that’s what the data are telling us).

So then the Chinese cats who coexisted with humans are a “false dawn” of domestication? Perhaps a dead end? If you view them in isolation, perhaps. But they shouldn’t be. Rather, they are another datum in a sequence of results which implies that the “Anthropocene” has been characterized by humans refashioning their environment so that a range of other organisms enter into coexistent, and sometimes domestic, relations with them. The domestication of dogs and cattle seem rather complex affairs, and may have had multiple starts, stops, and dry runs. The coexistence of some small (or, if you are a Drosophila geneticist, megafaunal) metazoans with humans as they began to co-opt a larger fraction of the earth’s biosphere should be no surprise. Instead of one single domestication it is possible that many organisms had multiple domestications, even if in many cases only a single event leaves a genetic traces today. That’s because domestication is a selective event, and works with the phylogenetic variation which is available to it. That variation is often wide enough that different lineages may develop “domestic” morphs rather readily. That may be what was happening in East Asia before the extant catus populations overwhelmed the nascent oranta derived domesticates.

Citation: Earliest evidence for commensal processes of cat domestication.

Why inbreeding is bad

A shocking case of a family of ~40 in rural Australia, the “Colts” (it’s a pseudonym), which has engaged in several generations of first degree incest has surfaced. You can read the summary in the press. But the Australian government has released a report on the case. I haven’t read most of it because the snippets I have stumbled upon are very disturbing. But, I was curious as to the characterization of the 12 children who were removed by social services. In particular, only one, Cindy, had parents who were unrelated. Note how different she is:

Cindy Colt (5), Rhonda Colt’s daughter, was medically examined on the day of her removal. She had a viral cold, but her health and hygiene was otherwise observed to be good, and her clothes were clean. She was suffering from an ear infection, although her mother had taken her to hospital two weeks prior to her removal to have this problem treated. She was unable to brush her teeth properly, though it is to be noted she was only 4 at the time. She could not bathe or dress herself, but unlike other children was reportedly capable of using toilet paper. But she preferred to eat with her fingers. She also required dental treatment, although it was submitted that her needs in this regard may not have been readily apparent to a lay observer, given the apparent absence of complaints of pain. Unlike the other children, Cindy presented as a well-spoken polite, bright, intelligent girl whose development was normal for her age. As previously noted, of all the children, the genetic testing demonstrated that her parents were not related.

From the descriptions it sounds as if most of the other children suffer at least mild retardation, and that factor compounded the clear neglect and abuse at the hands of adults. The lives of the children on the farm seem analogous to “Lord of the Flies,” with a large dosage of incest thrown into the mix.

The Great Felsenstein speaks

If there is one reason to read Larry Moran’s weblog, Sandwalk, is that Joe Felsenstein is a commenter there. There are two in response to the recent discussion over David Dobbs’ piece in Aeon Magazine. First, about Neo-Darwinism:

I think I can interpret what the first tweet refers to: the Modern Evolutionary Synthesis has been challenged by people who invoke epigenetics, phenomena of evo-devo, punctuationism, and a variety of other existent and nonexistent phenomena. the Synthesis has survived all these. It has turned out that some of these phenomena either do not exist (as in the case of neo-Lamarckian mechanisms). Other such as epigenetic mechanisms have little long-term effect. The rest can be considered to be examples of the workings of the mechanisms of mutation and selection, as in the case of evo-devo phenomena.

Similarly the Synthesis incorporated neutral mutation and nearly-neutral mutation. As for the word “neo-Darwinian”, it is gradually falling into disuse. This is partly for an historical reason: it turns out that in the late 1800s there were a group of scientists influenced by Darwin who championed the mechanism of natural selection when many other scientists were abandoning it, and they were referred to at the time as neo-Darwinians.

Pinker and Coyne are using “neo-Darwinian” simply as a synonym for the Modern Evolutionary Synthesis.

“Neo-Darwinian” is also sometimes taken as a modern-day name for panselectionist. That is what you are doing. Panselectionism has lost ground, but not the Modern Evolutionary Synthesis.

Then, in a post by Larry Moran on what I don’t like:

I’m not sure where Razib Khan is supposed to have opposed neutralism and random genetic drift. You seem to have decided that “evolution orthodoxy” is not wrong but is insufficient.

Genetic drift has been part of the Modern Evolutionary Synthesis since at least the 1920s — Wrights great papers from 1931 on were preceded by papers by R.A. Fisher that made use of genetic drift and by a paper by J.B.S. Haldane in 1927 that made use of it as well. The standard stochastic process model for genetic drift is called the Wright-Fisher model, and it dates from 1930 and 1931.

Neutrality was not seriously proposed as applying to actual genetic variation until Lewontin and Hubby’s 1966 paper (no, not in Crow and Kimura’s 1964 paper, which used it as a null case but did not suggest that this was real). While neutral mutation as a source of genetic variation and change was under active consideration from that point on, the idea that it refuted the Modern Synthesis was a non-starter.

Gould was an important evolutiuonary biologist, a masterful (if ultimately logorrheic) popularizer. He did much to put punctuational patterns of evolution on the map. He was joined in this by Raup, Sepkoski, Eldredge and Stanley. Their view that this required that most major change be due to species selection is not a majority view among evolutionary biologists — their argument did not gain widespread acceptance, though it still around.

The result of all this is that the horrible, boring, overly-orthodox Modern Synthesis has been refuted and swept into the dustbin of history — leaving us with the not-so-horrible, not-so-orthodox, but still boring Modern Synthesis.

The opinions of those who are eminent actors in the drama of science have greater weight. In these domains opinions make the transition to authority, whether the speakers wish it to or not.