Anthropology.net

Yesterday, my father emailed me a link to the Iranian Genome Project that caught my eye. Ironically, Razib over at Gene Expression also highlighted this project in a recent post. Much like the intentions Harappa & Dodecad ancestry projects, of which I’ve participated in by submitting my 23andme data, the Iranian Genome Project aims to enlighten Iranian heritage and health. As an Iranian American who follows population genetics regularly, I am very keen on intersection of these two topics.

I’ll be following the project, but honestly I don’t have high hopes. I would love to be proven wrong. It seems lofty, using a lot of high yield buzzwords. My first impression was if this nothing more than a CV booster … Especially since it hasn’t been updated since last September. I guess it can’t be completely an empty shell because they have an impressive member on research team, Pardis Sabeti.

You can learn more about this project by checking out their site, watching the following video and following them on Twitter: @irangenes. If you want, you can participate in the project by filling out this survey.

Inspired by the Dodecad Ancestry Project by Dienekes Pontikos and Eurogenes Ancestry Project by David Wesolowski, Zack Ajmal (with the help of Razib Khan) has started the Harappa Ancestry Project. Zack explains the motivation behind this project,

“It is a project to analyze (autosomal) genetic data of participants of South Asian origin for the purpose of providing detailed ancestry information. So the focus of the project is on South Asians: Indians, Pakistanis, Bangladeshis and Sri Lankans.

The project will collect 23andme raw genetic data from participants to better understand the ancestry relationships of different South Asian ethnicities.

I have named it after Harappa, an archaeological site of the Indus Valley Civilization in Punjab, Pakistan.”

There was a nice deal running on 23andme about a month ago for their ancestry & health kit that worked out to be $160 for 1 year. I hopped on board, got my kit, spat in the tube and sent it off. It is currently being analyzed. My ancestry is one of the populations Zack is looking for — so I’ll be sending my data to him. I can’t wait.

If you have had a 23andme genetic testing, you should consider participating in this project. It looks to be very interesting.

Both Razib and Dienekes have reviewed a paper on the population genetics of Central Asian peoples. To make sense of Central Asian ancestry has been challenging, to say the least. In particular, the problem is compounded by nomadic peoples without much written history nor uncovered archaeological record.

Ethnicities of Central Asia

What we do have are the linguistic, physical features, and now because of this paper some of the allelic traits of the different populations. Razib has pointed out some strange phrases from the paper that make me wonder about how much background on Central Asian cultures, migrations and phenotypes the authors really knew before publishing. There is really no confusion that more western Central Asian people look more western while more eastern Central Asian people look eastern, with some but little, shared traits.  But I don’t put total blame on them for not doing their research, it’s hard to make sense of the ancestry of Central Asia.

Razib has done a nice job explaining some of the previous cultures. Do check his post out. But a quick introduction for those who want to know, the steppes of Central Asia during the pre-Islamic periods, were predominated by sedentary Iranian peoples like the Sogdians, Chorasmians, Scythians, and Alans. Between the 5th-10th century, Turkic peoples moved from the east through the west. Turkic is a name given to a group of people who share a linguistic ancestry, Altaic. Some of these groups you may know are the Uyghur and Tatars. The Hun are possibly Turkic. Another major Altaic, but not Turkic, migration occurred with the Mongols during the 13th century.

There has been some confusion regarding the folklore and historical record compared to the phenotypic and linguistic differences on just how impactful the Turkic replacement been. The western historical record indicates that the invading Hun of the 5th centuries and Mongols later made a significant impact, wiping out large portions of ancient ethnic Iranian populations. This understanding is both true and false. There is evidence of entire cities being destroyed. At the same time, in texts like Ghenghis Khan and the Making of the Modern World, you read statements on how the Altaic invasions was much less of a violent horde and was demonized because of their comparative weakness in written language. In other words, the captors of the Mongolian Empire wrote their account of their overlords.

To this day, this has lead to “nationalistic” and ethnic conflicts and confusion, as evidenced by the June massacres of the Uzbeks by Kyrgyz groups regarding ancestry and heritage. The general consensus is the Tajik and Uzbeks were once a majority Indo-European-speaking population that were assimilated by migrating Turkic-speaking groups. The divergence from Middle Iranian to Turkic and New Persian was predominantly the result of an elite dominance process, as Razib points out.

So, just to which ancestry do Tajiks and Uzbeks, who share a Indo-Iranian language family and the Karakalpaks, Kazaks, and Turkmen, who share a Turkic language family belong to? With a 1,500 year shared regional history but linguistic separation, is it possible to flesh out if Turkic people invaded the West and replaced populations, or was there a back flow of Westerners who moved east?

Geographic-Linguistic-Genetic location of 26 Central Asian populations

The results from the paper out in The European Journal of Human Genetics, indicate that,

“The analysis of genetic variation reveals that Central Asian diversity is mainly shaped by linguistic affiliation, with Turkic-speaking populations forming a cluster more closely related to East-Asian populations and Indo-Iranian speakers forming a cluster closer to Western Eurasians. “

STRUCTURE plot on Central Asian Populations

Dieneke points out how the STRUCTURE plot (above) lets us see the that eastern Hazaras and Uyghurs have remained relatively separate from the more western peoples. Furthermore, supplemented by Razib’s comment,

“The eastern Turkic groups seem the least impacted by the Iranian substrate which was dominant before the arrival of Turks, while the Turcoman group sampled from western Uzbekistan seems to have been the most genetically “Iranized.”

…the correspondence analysis shows the Turkic groups exhibited a linear distribution toward East Asia, while the Iranian ones were placed where you’d expect them geographically.”

The data from this paper indicates that Turkic people did in fact move west, especially the men, since there’s high degree of  genetic homogeneity on the Y chromosomal lineage. They remained more genetic and linguistically unified and did not assimilate into Iranian genetics and languages. Additionally, contrary to popular belief, they did not absorb large populations of Iranians as their genetics and languages remained more separate than integrate.

A disclaimer, this is but one paper, with limitations on the number allelic markers that would make fine population differences more noticeable. But we can still see large trends regarding the ancestry of Central Asian people.

Martínez-Cruz B, Vitalis R, Ségurel L, Austerlitz F, Georges M, Théry S, Quintana-Murci L, Hegay T, Aldashev A, Nasyrova F, & Heyer E (2010). In the heartland of Eurasia: the multilocus genetic landscape of Central Asian populations. European journal of human genetics : EJHG PMID: 20823912

Related Articles

• Major study of Central Asian populations (Martinez-Cruz et al. 2010) (dienekes.blogspot.com)

Almost every biological anthropology text-book I’ve ever looked at has described the adaptations of human populations to the environments they occupy. Examples they give are the short stalky Inuit adapted to conserving heat in cold environments, the long lanky East African nomads adapted to far distant travels, and the barrel chested Peruvian and Tibetans living in low oxygen environments.

Highland Tibet

Little discussion, beyond correlating ecology and physical observation, is given to these. Actually I lie, the physiology of the barrel chested high altitude occupants is given a couple of sentences as well as an elevated oxygen binding capacity without concentrating their blood.

A paper published in Science several days ago tackles this latter issue. A group of scientists looked for unique alleles among Tibet highlanders and discovered 10 unique oxygen-processing alleles. I don’t have full access to the publication, so can’t tell if these genes encode for completely different functioning proteins or are differentially regulated at high altitudes.

All I can derive is that these genes seem to prevent polycythemia, edematous swelling of the lungs and brain, and hypertension of the pulmonary vasculature, which are all complications of high altitude living.  Two of these genes are EGLN1 and PPARA. PPARA is a peroxisome proliferation proteins that also is a leukotriene antagonist. That is interesting because in obstructive conditions like asthama, leukotrienes induce vasospasm and bronchconstriction. EGLN1 is also has an interesting role,

“it is a protein encoded by this gene catalyzes the post-translational formation of 4-hydroxyproline in hypoxia-inducible factor (HIF) alpha proteins. HIF is a transcriptional complex that plays a central role in mammalian oxygen homeostasis.”

These two genes were significantly associated with the decreased hemoglobin phenotype that is unique to this highland population.

Simonson TS, Yang Y, Huff CD, Yun H, Qin G, Witherspoon DJ, Bai Z, Lorenzo FR, Xing J, Jorde LB, Prchal JT, & Ge R (2010). Genetic Evidence for High-Altitude Adaptation in Tibet. Science (New York, N.Y.) PMID: 20466884

From Nobel Intent comes news of a discovery in the Mendelian genetics of Mirror Movements, a condition that causes people to involuntarily move both sides of their body when they intended to only move one.

Aside from being medically relevant, interesting on a population genetics level, and involved an Iranian family, it also caught my eye because about 3 weeks ago we covered the implications of DCC (deleted in colon cancer gene, I know — very clever!) mutations in my pathology course. DCC mutations are found in the sequence of events that lead up to a special type of familial adenomatous polyposis (FAP), known as Gardner syndromes.  These colon cancers occurs primarily on the left or descending colon. The morphology of FAP cancers lead to a napkin ring like constriction of the colon that present as alternating bouts of diarrhea and constipation. What makes them unique from other FAPs is that they have present with extracolonic manifestation, like bone cancers.

The DCC gene is on the long arm of chromosome 18. I know that it is a cell surface protein responsible for cell-to-cell and cell-to-matrix adhesion. Normally when cells proliferate, they squeeze up on each other and DCC works via contact inhibition to signal a stop in proliferation because conditions are getting too cramped. Therefore, if DCC is deleted, contact inhibition is lost and cell loses ability to proliferate, yielding a dysplastic growth.

Genbank classifies this gene as one that encodes for a netrin 1 receptor, which I did not know before I read this post. I find this really interesting in the relevance of DCC to Mirror Movements. Dr. John Nicholls,of SISSA in Trieste, Italy,  the dude for neurodevelopment, guest lectured my neuroscience course during my second term of medical school last year. I remember him describing netrins as a class of axon guiding proteins that functioned during growth and development. The hallmark experiment I remember him citing was the Oster, et al., 2004, where ganglion cell axon pathfinding in the retina and optic nerve was guided by netrin signals.

It seems that in Mirror Movements, the mutation in DCC prevents it from helping,

“nerve cells on one side of the spinal cord to stay on that side as they extend processes up and down the developing spine…. Because the protein is malformed, the body develops neural connections that route one-sided connections to both sides, producing the mirrored activity.”

I don’t have access to Science unfortunately to research the demographics of the particular SNP they discovered… So I can’t tell you of the gene frequencies… But if anyone does have access to the paper, and doesn’t emailing me, I’ll be very grateful. I love these sorts of discoveries where I learn something new and integrate what I’ve learned the past year and half of medical school!

Srour M, Rivière JB, Pham JM, Dubé MP, Girard S, Morin S, Dion PA, Asselin G, Rochefort D, Hince P, Diab S, Sharafaddinzadeh N, Chouinard S, Théoret H, Charron F, & Rouleau GA (2010). Mutations in DCC cause congenital mirror movements. Science (New York, N.Y.), 328 (5978) PMID: 20431009

Kambiz here. I’m about to start my second term of medical school, which is both exciting and nerve racking. In my summer readings, I came across a medical and anthropological tidbit today that caught my attention: redheads have a lower tolerance for pain. I didn’t know that.  Did you?

Skin pigmentation is one of my favorite topics. We know from previous posts that the melanocortin-1 receptor gene or MC1R affects melanin production and ultimately skin and hair phenotypes. Redheads carry a variant of MC1R which produces a different pigment, called pheomelanin, resulting in freckles, fair skin and ginger hair.

How MC1R receptors affect pain is another story though, one that is not well understood. Aside from the skin MC1R is also expressed in the brain. It could be possible that the redhead allelic variant of the MC1R receptors don’t quite receive the signal transduction of pain reception in the same manner as those with the wildtype receptor. I don’t think I’m gonna become an anthesthesiologist but as someone interested in human genetic variation it is good to acknowledge some phenotypes affect how medicine is delivered. ;-)

Anyways, I don’t have much else to add to this other than to share some interesting information and to let you all know I’m alive but will be crawling under a rock again. Till next time.

The Pritchard lab has put up an awesome new interface to query the data from the Human Genome Diversity Project, the HDGP Selection Browser. This is browser is phenomenal. You may have known about a previous iteration, Haplotter, also made by the Pritchard lab, which isn’t too user friendly and restricted to only data from four populations.

SLC24A5 SNP (rs2433354) Distribution Frequencies

The new HDGP Selection Browser integrates over 650,000 SNPs from 968 individuals originating from 52 different populations, making it much more granular data. As Razib also stated, the database is queried via GET, meaning we can hotlink to genes of interest, like SLC24A5: a gene related to skin coloration. The authors also provide a map to view how the certain alleles are geographically distributed. For example, see how this SNP, rs2433354 is spread throughout the world. Humans are really genetically different!

Daniel MacArthur, of Genetic Future, has a really awesome do it yourself post on how to use the database. I recommend you check it out if you’re interested in know how to figure out how populations vary on a gene to gene basis. Also, Daniel explains how to look to see if a certain gene or allele has population-specific selection. I won’t rehash and try to steal his thunder since he did such an excellent job.

I tip my hat to the Pritchard lab for developing such a fine database interface. I’ve been working on making my own database and it is not easy to make such a fluid and well executed application and user interface. You guys really made a stellar tool that I will be using a lot in the future.

One of the pieces to appear in the latest Science is Constance Holden’s synopsis of the core issues discussed at last week’s meeting of the National Human Genome Research Institute: defining geographic populations, handling interpretations of race (especially as as a sociopolitical term), and phrasing results of population genetic studies.

I paid cursory attention to the etymological aspects of the piece. Yes, I know Amerindian isn’t how some Native Americans want to be identified as, and there are some problems with figuring out where European populations end and where Asian populations begin. But I’m hopeful, as more individual genomes are sequenced and released, that genetic patterns can better define populations than cultural and geographic categories have in the past. We don’t necessarily have to rephrase terms or agree on new ones, but can possibly use biological terms, such as allele frequencies, as defining characteristics of populations.

Holden also reviews a discussion on interpretations of fitness — i.e. how some of the public may interpret Carlos Bustamante’s recent Nature paper, where he concluded that European-Americans had more deleterious gene mutations than African-Americans. Does that mean there’s some sort of superiority? No, but that doesn’t mean the public won’t interpret it like that. Should scientists hold back on their reporting their results or sugar coat them just to prevent the public from over analyzing them? I don’t think so.

The must read part of this news piece, especially for anyone news piece for anyone interested in the current state of population genetics and molecular anthropology, is the heated debate between Bruce Lahn and Celeste Condit, a professor of speech communication at the University of Georgia, Athens. Bruce Lahn, as you may know reported 4 years ago that selection in mutations of two genes (ASPM & microcephalin) regulating brain development is more common in Eurasians than in Africans. Condit argued that Lahn’s results have a political message embedded, a common mistake that many uneducated critics of population genetics repeat. We’ve had similiar misconceptions raised on Anthropology.net. Lahn retored back that some…

“are almost like creationists” in their unwillingness to acknowledge that the brain is not exempt from selection pressures.”

Oh snap! The whole meeting didn’t seem to be fruitless though, most agreed that suppressing freedom of reporting results as they are observed in the name of political correctness is not conduicive to the scientific method.

C. Holden (2008). PERSONAL GENOMICS: The Touchy Subject of ‘Race’ Science, 322 (5903), 839-839 DOI: 10.1126/science.322.5903.839a

Today’s issue of Nature has a brief essay on the role of language in cultural evolution. The authors touch up on a lot basics, such as anatomical localization of brain activity related to language and tool making, FOXP2, and how language has helped humans pass on cultural information more effectively than any other form of communication. Overall, it is a well written review that I want to pass on.

Related, Erin from the Spitton, shared news of the identification of a new language related SNP on the gene CNTNAP2. The paper which reports this is titled, “A Functional Genetic Link between Distinct Developmental Language Disorders,” and was published in the New England Journal of Medicine. I believe it is open access, I got to the full text with no problem. The authors hypothesized that neural pathways downstream of FOXP2 can also affect language impairment.

To identify possible downstream candidates that might be involved in typical SLI, the authors transfected a human brain cancer cell line (SH-SY5Y) to continually express FOXP2. FOXP2 is a transcription factor, meaning it is a controller of the expression of other genes. If it is mutated, it can’t regulate its targets properly and leads to different, sometimes mutant, phenotype. The used a type of test called the chromatin immunoprecipitation (ChIP) assay which identifies how and often where proteins, like the FOXP2 transcription factor, bind to specific regions of the genome. This is done by using specific antibodies that recognize a specific protein or a specific modification of a protein, in this situation anti-FOXP2 antibodies.

The ChIP assay showed that the FOXP2 transcription factor binds to a particular, novel region of interest, the first intron of gene CNTNAP2. When transcribed and translated, CNTNAP2 normally encodes for the protein CASPR2 — a protein that is localized and understood to function in the nodes of Ranvier on myelinated neurons. Of further interest, CNTNAP2 is expressed in the human cerebral cortex, specifically the orbital gyrus and superior frontal anlage, spanning the inferior and middle frontal gyri — all regions know to related to language cognition.

To make sure that FOXP2 was for sure targeting this region, and wasn’t mislead due to any conformational changes that came from the antibody it was complexed with, the authors did some PCR and sequencing and saw that this region of interest, intron 1, does have matching known consensus, binding sequence for FOXP2. They did some other tests that shows that this sequence is highly specific to FOXP2… all of which suggests that this site on CNTNAP2 is definitively a binding site for FOXP2 (CAAATT).

The authors next varied the amount of FOXP2 expression and tried to see if it affects the ultimate expression of CNTNAP2. They were able to show there is a correlation — CNTNAP2 transcript levels were lowest where there are higher levels of FOXP2, suggesting that FOXP2 down regulates CNTNAP2. We haven’t know about FOXP2-CNTNAP2 interactions before, because FOXP2-bound fragment of CNTNAP2 is outside of the classically defined regulatory regions that promoter based microarrays identify… So identifying this pathway is very commendable.

With this downstream candidate gene isolated the authors moved to see how polymorphisms in CNTNAP2 manifest language phenotypes. Their population sample was made up from children from 184 different families where at least one child had a specific language impairment (SLI). The children had wildtype FOXP2, but children who carried the guanine nucleotide at rs17236239 SNP on CNTNAP2 had worse scores on a test that measures their ability to reproduce nonsense words like “brufid” and “contramponist.”

Now don’t get me wrong, this SNP, rs17236239, ain’t on intron 1 — where FOXP2 binds. FOXP2 was used as bait to fish out what gene bites to it. When CNTNAP2 was figured out to be a new novel target of FOXP2, the authors tried to see if CNTNAP2 variations also affect language. And they do. What’s also of interest is that other SNPs in the same regaion that rs17236239 is found also have CNTNAP2 as been linked to delayed speech in children with autism.

I’m really impressed with this paper. It’s a gem. Well written and straight forward. I don’t regularly read papers of such caliber, to be honest… So I really appreciate when I do. The new language related gene is also very important as we begin to piece together the complex network of genes and proteins, anatomy and behaviors that have allowed us to have language and use it.

Eörs Szathmáry, Szabolcs Számadó (2008). Being Human: Language: a social history of words Nature, 456 (7218), 40-41 DOI: 10.1038/456040a

S. C. Vernes, D. F. Newbury, B. S. Abrahams, L. Winchester, J. Nicod, M. Groszer, M. Alarcon, P. L. Oliver, K. E. Davies, D. H. Geschwind, A. P. Monaco, S. E. Fisher (2008). A Functional Genetic Link between Distinct Developmental Language Disorders New England Journal of Medicine DOI: 10.1056/NEJMoa0802828

If you’re a regular reader of Dienekes blog, you’d know he’s consistently raised concerns that calibrations of molecular clocks don’t quite fit the bill. Yesterday, he posted an addendum and shared a new paper in which authors advocate that molecular clock can be calibrated upon an archaeological context (not phylogeny-based) and human mtDNA estimates of dates of population and phylogenetic events should be adjusted to time-dependent mutation rate estimates.

I’m not gonna get into a rehashing of Dienekes’ post, I wouldn’t do as good of a job even if I did… but you should jump on over and read what he has to say and how he explains his criticisms of how the clock has been calibrated in the past. I want to spend some time in this post discussing some of the results of the paper he shared, “Characterizing the Time-Dependency of Human Mitochondrial DNA Mutation Rate Estimates,” in Molecular Biology and Evolution. The authors sought to establish genealogy-based estimates of the mtDNA mutation rate using both hypervariable and coding region data, they also wanted to figure out if multiple hits  affect the discrepancy between the different methods of mutation rate estimation.

So they setup new genealogy-based rates from 2,500 to 50,000 years ago using mtDNA from populations in the Canary Islands, Polynesia, Micronesia, North America, Taiwan, Indonesia, and Oceania. The populations were selected based upon relative isolation and the  available archaeological dates for the time of first human arrival, haplotypic data from neighboring regions, and indigenous haplotypes for that region.

The authors were able to calculate that the evolutionary mutation rate between approximately 2,500 and 50,000 years ago was much different than that from 50,000 to 6 million years ago. They suggest that since earlier mutation rates, ones based upon pedigrees, are not affected by the processes of
bottlenecks and selection, except for purifying selection on lethal alleles, they can’t weed out the effects demographic processes. Using their time-dependent approach they observe that molecular clock was accelerated for large Neolithic populations and is similar to the pedigree rate, but for the smaller Paleolithic hunter-gatherers it was much lower…. makes sense, as populations grow, variability accelerates.

B. M. Henn, C. R. Gignoux, M. W. Feldman, J. L. Mountain (2008). Characterizing the Time-Dependency of Human Mitochondrial DNA Mutation Rate Estimates Molecular Biology and Evolution DOI: 10.1093/molbev/msn244