Anthropology.net

A new paper in the open access journal PLoS Genetics reports on a comparison of genetic, geographic, and linguistic patterns of the diverse populations found on the major islands of the Bismarck Archipelago and Bougainville, Melanesia. The paper is titled, “Genetic and Linguistic Coevolution in Northern Island Melanesia.” I think that Simon Greenhill of HENRY may know a bit more about these populations, languages and region than I, but I’m gonna still try and summarize the paper and briefly discuss the results.

The earliest inhabitants of the area arrived around 40,000 years ago, but there was an additional migration into the region about 3,300 years ago. We know that primarily because of the linguistic diversity. The two major languages are Oceanic and Papuan. Oceanic, being a major branch of the widespread Austronesian language family, and the Paupuan languages, likely descendants of languages spoken by people who began arriving in the region more than 40,000 years ago. The rugged geography of the region has been a cause for a lot of the diversification. Despite their regional affinities, the two languages do not form a very coherent language family.

Genetic, Geographic & Linguistic Distances Of Melanesian Popultions

The study sampled 776 individuals from 33 linguistically based populations, which averages to about 23 individuals per population. Each individual was typed on 751 different autosomal microsatellites. The languages were compared on 108 different structural linguistic features. The authors applied two different tests to figure out if genetic and linguistic similarities were formed following early population splits and isolations or if the genetic and linguistic similarities were formed through continuing genetic and linguistic exchange between neighboring populations.

The authors were able to figure out that genes moved freely than languages between nearby populations, regardless of the language family (compare Figures B to D). Language exchanges, on the other hand, have been particularly limited between neighboring Oceanic and Papuan languages (check out Figure D & F). In certain regions, like the rugged interior of the largest island, New Britain, the authors found strong correlations between genetic, linguistic, and geographic distances when compared to their less restricted coastal neighboring populations. They are almost always distinctly different. While extremely restricted to several islands, this study shows us a scenario where language barriers do not particularly hinder genetic exchange, but geography still does.

Keith Hunley, Michael Dunn, Eva Lindström, Ger Reesink, Angela Terrill, Meghan E. Healy, George Koki, Françoise R. Friedlaender, Jonathan S. Friedlaender (2008). Genetic and Linguistic Coevolution in Northern Island Melanesia PLoS Genetics, 4 (10) DOI: 10.1371/journal.pgen.1000239

The Phoenician Alphabet

The Phoenician civilization is understood to be the dominant maritime trading culture between the period of 1550 BC to 300 BC. While they were based out of the Levant, their city-states were spread all across the Mediterranean. The golden age of Phoenician culture and seapower is usually placed around 1200–800 BC. When Cyrus the Great conquered Phoenicia in 539 BC, he divided the Phoenicians into four vassal kingdoms by the Persians: Sidon, Tyre, Arwad, and Byblos. Each flourished, building fleets for the Persians against the Greeks. But their autonomy as distinctly Phoenician people declined after this. The lasting and most important cultural legacy of Phoenicians on modernity is their alphabet. It is generally thought that their alphabet is the ancestor of most modern alphabets.

Okay enough of a history lesson, a team of researchers developed a set of algorithms to detect the subtle genetic impact of historical population migrations. They’ve tested out their formulas on 1,330 men in hopes that they’ll be able reveal the genetic legacy of the Phoenicians. Specifically, they have made a new set of tests that seek out patterns in genetic signatures of modern men. They’ve published their research in the the American Journal of Human Genetics under the title, “Identifying Genetic Traces of Historical Expansions: Phoenician Footprints in the Mediterranean.”

The team sampled Y chromosomes of men from historic Phoenician trading centers in the Mediterranean regions of Syria, Palestine, Tunisia, Morocco, Cyprus, and Malta. After genotyping them, they compared them on 11 STRs and 58 Y-SNPs markers. They weeded out background variation from previous Neolithic migrations, and singled out more widespread Greek colonization events from isolated Phoenician expansions, such as the Phoenician colonization of Tunisia.

The Phoenician Genetic Footprint In the Mediterrenean

The authors were able to detect a half dozen haplotypes and they call them Phoenician Colonization Signals (PCS). PCS3+ is calculated to be the strongest Phoenician-colonization candidate. It is tightly associated with the SNP haplogroup E3b, but it does not show the wide geographic coverage that the other PCS+s demonstrate. Both PCS1+ and PCS2+ score well, although not as strongly as PCS3+. The excess of haplogroup J2, and PC1+ to PS3+ in coastal Tunisia, the site of Carthage, compared to inland Tunisian populations is exceptionally significant, and suggests that the Roman destruction of Carthage did not eliminate the Carthaginian gene pool. So the presence of these seven related genetic lineages in places around the Mediterranean Sea, tell us that where Phoenicians had lived and persisted genetically.

These lineages suggest that the Phoenicians contributed their genes to at least six percent of modern populations of historic Phoenician trading outposts. In fact, one boy in each school class from Cyprus to Tunis may be a direct male-line descendant of the Phoenician traders.

Of course, since this is only a Y-chromosome test, we’re only getting part of the genealogical history. If a Phoenician man fathers only daughters, his Y-chromosome lineage dies out. That means tests likes these can only say something when there’s an unbroken male line in that area. It is certainly possible that more people from Cyprus to Tunis have a Phoenician heritage. Dienekes, a Greek, has a scathing criticism of the paper. This paper explicitly says they didn’t try to seek out Greek expansion but Dienekes outlines six shortcomings, related to Greek expansions, that the paper didn’t factor that would affect these conclusions — he ends his post saying,

“Is there anything of value in this paper? Well, it’s a good idea to try to correlate Y-chromosome distribution with historical rather than pre-historical events. Too bad the authors botched the job, but their paper can at least serve as a reference point for how not to go about doing it.”

Pierre A. Zalloua, Daniel E. Platt, Mirvat El Sibai, Jade Khalife, Nadine Makhoul, Marc Haber, Yali Xue, Hassan Izaabel, Elena Bosch, Susan M. Adams, Eduardo Arroyo, Ana María López-Parra, Mercedes Aler, Antònia Picornell, Misericordia Ramon, Mark A. Jobling, David Comas, Jaume Bertranpetit, R. Spencer Wells, Chris Tyler-Smith, The Genographic Consortium (2008) American Journal of Human Genetics. DOI: 10.1016/j.ajhg.2008.10.012

Last month I was excited to share some research about the chemical composition of Ötzi, the 5,000 year old Tyrolean Iceman that has captured my attention for quite sometime. Today, I’m even more excited to share that the complete mitochondrial genome of Ötzi has been sequenced using a combination of PCR amplification and 454 sequencing. The research has been published in Current Biology. You can find it under the title, “Complete Mitochondrial Genome Sequence of the Tyrolean Iceman.”

I’ve covered the details behind Ötzi before. I’ll give you a quick run down in case you forgot or never knew about him. Ötzi is the name given to mummy discovered on September 19^th, 1991, around 3,270m above sea level, in the Eastern Alps near the Austro-Italian border. His remains were dated to be , 5,350–5,100 years old, and was remarkably preserved because of the cold climate. We have an idea what his last meal was and what he wore. He’s thought to have been around 46 years old, his fertility has been questioned, and his cause of death seems to have been rather horrific — severely wounded by an arrow and some blunt force trauma to his face.

Previous researchers have sequenced some of his mitochondrial genome, specifically the hypervariable segment (HVS-I). Two nucleotide transitions, at positions 16224 and 16311, indicate that Ötzi’s mtDNA belonged to haplogroup K, a subclade of the major west Eurasian haplogroup U. The authors of the new Current Biology paper decided to completely sequence the mitochondrial genome of Ötzi using 454 pyrosequencing technology. They’ve compared the sequence to 115 published complete mtDNA sequences from modern individuals, and constructed a phylogeny of the K haplogroup.

The sequencing run seems to have been rather uneventful. I’ve covered 454 technology before, but to recap it is sequencing by synthesis, which involves template DNA being immobilized, and solutions of each nucleotide added. They hybridize to their complement at the first unpaired base of the template. The hybridization reaction lets off light, because the polymerase enzyme is paired with another other chemiluminescent enzyme. A high resolution photo is taken and any remaining unbound nucleotide is removed and then another wash of another base is made. With the array approach developed by 454, it is possible to generate over 100 million nucleotide data in a 7 hour run with a single machine.

The authors reported they made 45,829 reads, but only 42,695 reads, or 93.2% of the total, were usable. Several gaps in the mitochondrial genome were observed, so PCR products were cloned into vectors and sequenced with conventional Sanger technology. When compared with the revised Cambridge Reference Sequence (rCRS), the consensus sequence showed 30 mtDNA transitions. A phylogenetic comparison was made to all 115 haplogroup K complete sequences currently available. The authors confirm that Iceman’s sequence falls within haplogroup K.

But, transitions at positions 3513 and 8137 on the Iceman’s mitochondrial genome indicate that his maternal lineage belongs a K1 subhaplogroup but not to any of the three subclades into which K1 is currently further subdivided (K1a, K1b, and K1c). The authors conclude that the Iceman’s mtDNA, belong to a novel branch of K1, not yet identified before. They’re calling it K1ö.

I’m not gonna get into much of a discussion about contamination because the samples were taken from thawed tissue from the mummy’s rectum. Many lines of evidence show that there was a lot of endogenous mtDNA that woulda muddled out any contaminating DNA and the results reconfirm the previous HVS-1 results. Anyways, this is the oldest complete H. sapiens mtDNA genome generated to date. The results show that as the frequency of genetic lineages change over time, due to genetic drift, some variants die out. Based upon the mtDNA, it is highly unlikely that Ötzi has any modern day maternal relatives… unless we sequence more than 115 haplogroup K carriers.

Luca Ermini, Cristina Olivieri, Ermanno Rizzi, Giorgio Corti, Raoul Bonnal, Pedro Soares, Stefania Luciani, Isolina Marota, Gianluca De Bellis, Martin B. Richards, Franco Rollo (2008). “Complete Mitochondrial Genome Sequence of the Tyrolean Iceman” Current Biology, DOI: 10.1016/j.cub.2008.09.028

When the headlines behind this story came across my RSS reader the other day, I was gonna file it under my proverbial “Captain Obvious” category. The basic premise is the potential link between last name and Y chromosome type. We already know that in deep ancestries, like among Jewish people, the Cohen Modal Haplotype (CMH) is notably frequent amongst Cohens. Cohens are a lineage of Jews believed to be direct relatives to the Biblical Aaron. They have a relatively strict marital tradition, one where the last name is preserved paternally.

Many other cultures are structured paternally and last names are inherited, for the most part, from the father’s family. Turi King made this otherwise obvious observation and tried to look for similar Y-chromosome haplotypes among two and half thousand men with 500 different last names. She presented her results to the Doctoral Inaugural Lectures being held in the Frank and Katherine May Lecture Theatre, Henry Wellcome Building, University of Leicester last week.

While not as deep as Cohens, there should be a linkage with almost any man’s last name and his Y-chromosome genetics… So long as name changes, adoptions and multiple foundations of the same last name by different individuals (like Smith) haven’t been prevalent. And that’s why I changed my mind about posting this research. King explains,

“The last name Smith is a good example of this as it derives from the occupation of blacksmith so many men could have taken on the last name Smith. This means that instead of just one type of Y chromosome being associated with a last name, many different types of Y chromosomes would be associated with this single last name. On the other hand, for rarer names, there may have been just one founder for the name and potentially all men who bear that last name today would be descended from him and could be connected into one large family tree.”

When King compared the Y-chromosome makeup of non-related individuals limited down to 40 last names, she was able to see that last names like Attenborough and Swindlehurst showed that over 70% of the men shared the same or near identical Y chromosome types whereas last names such as Revis, Wadsworth and Jefferson show more than one group of men sharing common ancestry but unrelated to other groups.

Ultimately, King was able to show between two men that share the same last name, there is a 24% chance of sharing a common ancestor through that name but that this increases to nearly 50% when their last name is rare. For forensic scientists out there, especially forensic anthropologists who regularly deal with skeletal remains, this can be a godsend because a prediction of a male’s last name is potentially possible from DNA alone. Now disclaimer aside, the press release didn’t really mention her methodology, i.e. how many loci she compared. Of course, the more the merrier. Either way, I’m sure many out there will look forward to larger comparisons of last-name to Y-chromosome correlations.

Earlier this year, I wrote a massive summary on the genetics of pigmentation for one of my graduate courses. I wasn’t particularly keen on the topic before but it has since grown on me and I’m now a big fan. So to read from Yann, Dienekes, and Razib that one of the key pigmentation genes, SLC45A2 is involved in hair color among Polish people, I was both excited and, well, frankly not very surprised.

Before I drag you into a review of the paper let me first introduce the gene and its function. SLC45A2 stands for solute carrier family 45, member 2. It is found on the short arm of chromosome 5. Not very informational, but the protein it ultimately encodes for is known as membrane-associated transporter protein (MATP). As the name implies, this protein is thought to regulate traffic melanosomal proteins into melanosomes, organelles within pigmentation cells (melanocytes) where melanin is produced. Melanin functions as a protective agent. It is dark in color and accumulates in cells in reaction to sunlight, absorbing light and protecting the nuclear genome from mutations caused by the ionizing radiation from UV rays. Thus, making MATP and the genetics of SLC45A2 an important part of the skin pigmentation pathway.

Recent studies have shown that certain variants of MATP affect pigmentation, such as Cook et al., 2008 and Graf et al., 2007. In the latest piece that everyone is buzzing about, “Association of the SLC45A2 gene with physiological human hair colour variation,” the authors were able to make an significant association between one of two non-synonymous polymorphisms in SLC45A2 and the hair color phenotype of a Polish population.

Inspired by Razib's example: Marzena Cieslik, Miss Poland 2006

The two SNPs are rs26722 and rs16891982. Both are missense mutations that encode for a different amino acids. The first SNP, rs26722, is a change from a guanine base to an adenine at position 907 of the mRNA transcript. This swap to an adenine affects the resultant the codon, creating a lysine on position 272 of the amino acid sequence instead of a glutamic acid residue (annotated as E373K). Similarly, rs16891982, is also a switch of a guanine but for a cytosine base on a different position of the transcript, 1214. The leucine residue on position 374 is thus changed to a phenylalanine (annotated as L374F).

Biochemistry aside, these two end products, MATP-E272K and MATP-L374F, have distinct population frequency distributions, especially MATP-L374F. Among the Polish sample of 392 individuals of varying skin, hair, and eye color, the rare allele is L374… Found in only 2.3% of the population. The MATP-374F allele is represented in 97.7% of the Polish sample. Curiously, all people who carried the rare allele had dark hair. The authors calculated the odds the L374 genotype increased the likelihood a person would have dark hair by 7 times.

Razib linked up an awesome resource, called ALFRED, the allele frequency database which shows the distribution of rs16891982 world wide. You can see that in northern Europe, 374F is the prevalent allele. Only when you get to Italy, Spain, Turkey, do you begin to see more of the L374 allele (the dark hair allele).

Worldwide Distribution of the rs16891982 SLC45A2 Allele

This all makes sense, people who carry a cytosine on position 1214 of the SLC45A2 mRNA have lighter color hair than those who carry a guanine (L374). A similar distribution frequency was observed a German and Japanese sample by Nakayama et al., 2002 & Yuasa et al., 2004 & Yuasa et al., 2006. 96.5% of Germans had the lighter 374F allele. 100% of Japanese had the darker L374.

But why?

I returned to one of the previous studies I mentioned earlier, the Cook et al. paper, “Analysis of Cultured Human Melanocytes Based on Polymorphisms within the SLC45A2/MATP, SLC24A5/NCKX5, and OCA2/P Loci.” The authors of this paper grew primary melanocytic cells with specific mutations and tried to see the effects they had on melanin content and tyrosinase activity. In their array, one of their comparisons included the impacts of the MATP-L374 variant to the MATP-374F variant. They found that MATP-L374 cells expressed significantly lower MATP transcript levels compared to MATP-374F ones.

I’ve spent a lot of time thinking about this and quite frankly, I’m very confused. Correct me if I’m wrong, but lower levels of MATP transcript means less transport proteins translated and available to bring melanosomal proteins into the cells. Less building blocks equals less melanin. And less melanin should equal lighter pigmentation. So why do people who have dark hair have the less productive allele?

Come to think of it, there could be many different reasons, actually. Biochemical pathways are hardly ever a 1:1 mechanism. There’s lots of redundancy and sometimes one component isn’t the point man. For example, other parts of the pathway, such as tyrosinase, an enzyme that catalyzes the production of melanin, is higher in darker skin. Other transport proteins, such as SLC24A5/NCKX5 are also part of the melanin and melanosome production network and could have a different impact.

Either way, there are a few curious things to come out of all of this… We now know a definitive allele that affects hair color. Albeit, how is still up in the air, ’cause darker haired people come with less membrane-associated transporter proteins on their melanocytes. But still, there is a distinct structure to the distributions of these haplotypes.

Wojciech Branicki, Urszula Brudnik, Jolanta Draus-Barini, Tomasz Kupiec, Anna Wojas-Pelc (2008). Association of the SLC45A2 gene with physiological human hair colour variation Journal of Human Genetics DOI: 10.1007/s10038-008-0338-3

The other day Dienekes pointed out a paper on ancestral human population dynamics within Africa before the out of Africa migrations. The paper is very similar to one I reviewed in April, which also focuses on the diversity of the mitochondrial haplogroup L — one of the oldest mtDNA haplogroups out there.

The new paper, “Bayesian coalescent inference of major human mitochondrial DNA haplogroup expansions in Africa,” published in the Proceedings of the Royal Society B, uses coalescent theory to investigate past population sizes of each of the four major African mtDNA haplogroups (L0-L3). 224 different mitochondrial genomes were analyzed and the comparison yielded some similar results to the previous paper I mentioned. But remember, the last paper investigated the time of the emergence of each haplogroup. This paper focuses on effective population sizes.

Anyways, for starters, the results show that three distinct demographic histories can be seen from the underlying the four haplogroups. Two of the oldest haplogroups, L0 and L1, show exponential growth from 213,000 to 156,000 years ago. The previous paper suggested that the L0 and L1 split about 200,000 years ago. Soon after this split, one of the the paleoafrican branches L0 established what we now consider sub-Saharan Khoisan peoples.

L1 split up into the L2 and L3 branches sometime around 127,000 to 72,000 years ago, again consistent with the previous paper. The L2 and L3 branches show two exponential growth periods, one around 86,000 to 61,000 years ago and another around 20,000 to 12,000 years ago. The authors observed a distinct expansion of the L3 branch around 12,000 to 8,000 years ago. They suggest that,

“L3 did not simply spill over into Eurasia, but was driven as part of an expansion that had begun in sub-Saharan Africa thousands of years earlier.”

While this date is a bit later than the one suggested in the previous paper, both indicate that there were deep African migrations within Africa. The later expansions of L2 and L3, coincide with environmental and cultural changes, such as the greening of the Sahara and emergence of pastoralism. The authors write that,

“The timing of the L3 expansion-8-12kyr prior to the emergence of the first non-African mtDNA lineages-together with high L3 diversity in eastern Africa, strongly supports the proposal that the human exodus from Africa and subsequent colonization of the globe was prefaced by a major expansion within Africa, perhaps driven by some form of cultural innovation.”

Quentin D. Atkinson, Russell D. Gray, Alexei J. Drummond (2008). Bayesian coalescent inference of major human mitochondrial DNA haplogroup expansions in Africa Proceedings of the Royal Society B: Biological Sciences, -1 (-1), -1–1 DOI: 10.1098/rspb.2008.0785

As you know, we have many symbiotic relationships with bacteria. In fact, the average human gut contains about 1 kilogram (2.2 pounds) of these microscopic cells, some of which are beneficial. If you’ve ever had to take antibiotics, you may have come to learn how dependent we are on certain strains of bacteria to process plant materials… albeit through a rather uncomfortable experience. Certain flora, like E. coli, live in our intestines and help us synthesize vitamins K and B-complex, which are then absorbed by the body.

But microbes provide us with more than just dietary aide. Actually, microbes offer a unique line of evidence into investigating human migrations, which is crucial to have in anthropology. Since they evolve faster due to their exponentially expedited life span, we can trace how human population diverged from one another. One of my favorite papers on the subject was last year’s support for the Out of Africa model of human migrations using the genetics of Helicobacter pylori, another gut dwelling bacterial species.

Helicobacter pylori

A similar study on the topic was published today in the open access journal PLoS One. The title, “Amerindian Helicobacter pylori Strains Go Extinct, as European Strains Expand Their Host Range,” does a great job at explaining the main results of the paper. But I’ll try to provide a bit more detail in this post.

First, let me introduce the microbe to you. Unlike some of the other flora that reside inside of us, H. pylori, is more or less a parasite and a nasty one at that — known to cause stomach and intestinal cancers because of the chronic inflammation it causes. The inflammation stems off of the reaction of a component of its gram-negative staining walls, a lipopolysaccharide that initiates an immune response. However, most of us play host to this species and do a good job at keeping it under control. Inflammation only becomes a problem when combined with other deficiencies in the health and immune system of the individual.

The authors of this study made 131 bacterial cultures, 19 of which came from from Africans, 36 from Spanish, 11 from Koreans, 43 from Amerindians and 22 from South American Mestizos. They lysed the cells from the cultures and did lots of PCR and sequencing reactions for 7 housekeeping genes from each of the 131 cultures. When they did a comparison of the sequences they noted an interesting, but consistent result: decreased diversity of H. pylori strains in the human group with the least genomic diversity.

In other words, Native American human populations, who experienced a population bottleneck while crossing Beringia, also inadvertently caused a H. pylori bottleneck! We’ve suspected that Native American populations have undergone a bottleneck because looking at their ABO blood group diversity, they show a remarkable dominance of O blood type — something not seen in European and African populations.

Among Mestizo people, H. pylori genetic diversity was high. That’s cause Mestizos are people of mixed ancestry — a hodgepodge of Spanish and Native American hybrids. In fact, the authors could see the hybridization in the H. pylori from these people. Although bacteria reproduce asexually, they have this phenomenon, called recombination, which is effectively a sharing of parts of DNA. Mestizo H. pylori had some signatures from Amerindian H. pylori and some from European H. pylori indicating their was a sharing of H. pylori genes from these two different populations, just as there was a sharing of human genes.

Even though I’ve made this point before, similar to languages, material culture, and our own genetics, the genetics of organisms associated with humans, be it bacteria, rats, etc. can effectively tell us a lot about human migrations. This paper documents how a founder effect of sorts, a decrease of genetic variation of H. pylori among Native Americans due the effects of crossing Beringia some 15,000 years ago. It also documents the increase in genetic diversity of H. pylori as these Native American groups mixed with European ones.

Maria G. Domínguez-Bello, Maria E. Pérez, Maria C. Bortolini, Francisco M. Salzano, Luis R. Pericchi, Orlisbeth Zambrano-Guzmán, Bodo Linz, Angus Buckling (2008). Amerindian Helicobacter pylori Strains Go Extinct, as European Strains Expand Their Host Range PLoS ONE, 3 (10) DOI: 10.1371/journal.pone.0003307

We know that during the Sui dynasty, the Chinese empire had European residents. But what can be said about the diversity of China during a preceding dynasty, such as Qin Shi Huang’s empire — the Qin dynasty of China? A team of Chinese academics have analyzed the mtDNA of 19 individuals excavated from a nearby tomb at the Terra Cotta Army site. They have published their study in the open access journal PLoS One under the title, “Mitochondrial DNA Evidence for a Diversified Origin of Workers Building Mausoleum for First Emperor of China.”

The Terra Cotta Army Site in China

The remains date to 2,200 years ago, right around the time that Qin Shi Huang or Yíng Zhèng, was undertaking his massive project like building the Great Wall and the giant Terra Cotta warrior mausoleum. Yíng Zhèng is known for doing some great things for China. For example, he built an intricate road system which connected the twenty-two million inhabitants at the time and allowed him to control and unify such a vast territory. However, with such power came some questionable political measures. Zhèng buried alive many Confucian scholars and burned their books. And his large projects came at the expense of many people’s lives.

Although 121 skeletons were excavated from the Terra Cotta Army site, only 50 were genetically analyzed for this study, 19 of which yielded results. But based on morphological observations, all of them were robust and had signs of arthritis. Some had broken bones or signs of extensive muscular stress. All of this suggests that these people were engaged in heavy work before death… most likely slaves of the Emperor, forced to work on building such a massive mausoleum.

Razib raised issue that such ancient DNA analysis is subject to contamination, especially when trying to assess the genetic diversity of an ancient populations. But the authors seemed to have done the same ancient DNA song and dance that we’ve seen in the last year. They removed the outer layer of bone which was presumably handled under non sterile conditions. The bone was subject to UV radiation and chemical treatment to further nuke any exogenous DNA.

The extracted DNA was amplified via PCR with overlapping primers and TA sub-cloned. The samples were sequenced via the standard dideoxy-chain terminated method. Another independent lab preformed the same steps to replicate results and ensure contamination wasn’t much of an issue.

The ancient sequences were compared to 2,164 mtDNA profiles from 32 different Chinese populations. That was done to establish the geographic locations and ethnic affiliation of these ancient peoples to their modern contemporaries. Using the same primers, the mitochondrial genomes of all staff members that handled the remains, were also sequenced and compared. Because of the high amount of genetic diversity present in the results of the ancient DNA, and the high amount of homogeneity of the modern staff members’ mtDNA, the authors ruled out contamination from handling as a major issue.

Specifically, the 19 individuals came from 15 distinct east-Eurasia haplogroups. Overall, 4 of them were of Han origin. While 7 of them came from southern China. Three others came from minority groups in the south. Interestingly, one of these ancient workers carried the same variation in the haplogroup M7a seen in Ryukyuan and Japanese people.

Of course, as Razib mentioned, this sample size is small — only 19 of the assumed 720,000 workers it took to construct the mausoleum but this shows us that Yíng Zhèng recruited people from all over to work on his project. Is this surprising? No! It isn’t. We’re talking about an empire. Empires are huge, some span entire continents and are made up of many different elasticities. Other contemporaneous empires, such as the Romans and Persians, even the Egyptians, were hardly homogeneous. Why would the Qin Empire be otherwise? Especially under the control of the unifier, Yíng Zhèng, who pushed for the infastructure that connected China at the time.

Zhi Xu, Fan Zhang, Bosong Xu, Jingze Tan, Shilin Li, Chunxiang Li, Hui Zhou, Hong Zhu, Jun Zhang, Qingbo Duan, Li Jin, Vincent Macaulay (2008). Mitochondrial DNA Evidence for a Diversified Origin of Workers Building Mausoleum for First Emperor of China PLoS ONE, 3 (10) DOI: 10.1371/journal.pone.0003275

This morning Dienekes pointed out a new paper in the open access journal PLoS Genetics on polygyny and its impact on human genetic variation. Razib followed suite, providing a more in depth review of the study. I recommend you check out both. In this post, I’m also gonna have a stab at reviewing the paper since it has an important anthropological impact.

The paper, “Sex-Biased Evolutionary Forces Shape Genomic Patterns of Human Diversity,” is authored by some people you may have heard of, such as Michael Hammer and Jeffrey Wall among others. Like many population geneticists, they isolated the problems and limitations of previous studies which investigated genetic diversity of humans from only markers on the mitochondrial DNA and non-recombining portion of the Y chromosome genome. They proposed that markers on the autosomal genome, including the X sex chromosome, will provide a more insightful understanding.

So they compared the genetic variation among 40 independent loci on the X chromosome and autosomes in 90 individuals from six different populations. 20 loci on the X chromosome and 20 on the autosomes were picked from non-coding regions of the genome.

Why wasn’t there a more even distribution of sites across the whole genome? Well, the authors specifically sought to seek out the impact of sex-specific processes, such as mating patterns, in shaping genomic patterns of variability. Both Razib and Dienekes do an excellent job in explaining this, but I’ll snip what Razib wrote since it is more clear in my mind:

“Assuming equal numbers of males and females in any given generation you expected a ratio of diversity of 0.75 between the X and the autosomes; remember that the number of copies of X circulating within the population are reduced by 25% because males carry only one copy, while women carry two.”

In other words, the X chromosome is present in two copies in females and a single copy in males. We all know that. We expect that the other chromosomes will show more genetic diversity than the X chromosome in a population with an equal number of breeding males and females because they are inherited equally by both sexes from each parent. In a populations with an unequal number of breeding males to females, we should see something different. Actually, we expect to see more genetic diversity on the X chromosome than on the other chromosomes in areas where men don’t get to pass on their genes, while most women do.

The authors’ samples included individuals from Africa, such as the Biaka of the Central African Republic, the Mandenka from Senegal, and the San from Namibia were included. Outside of Africa, the French Basque, the Han Chinese and Melanesians were also sampled. Roughly 210kb of DNA was sequenced from each of these individuals, and a basic statistical summary of the nucleotide diversity in six human populations was conducted. Comparing the observed nucleotide diversity on the X chromosome to the chromosomes showed that there was more genetic differences in the X chromosome than would be expected if equal numbers of males and females tended to mate.

Even though I explained this in two paragraphs above, polygyny could be the only reason why we see such results. Some men just didn’t get a chance to pass on their genes. The authors even made sure to rule out background selection, changes in population size and sex-specific migration in their conclusion. Only the process of polygyny could account for the sex ratio skew and resulting patterns of genomic variation. By this process, fewer unique male genes are being passed into the next generation.

In the same issue, a very similar paper was also published that I don’t think many other people noticed. A separate team of academics applied this multilocus approach to the genetic diversity of Central Asia. It is published under the title, “Sex-Specific Genetic Structure and Social Organization in Central Asia: Insights from a Multi-Locus Study.” Their sample included 10 populations of bilineal agriculturalists and 11 populations of patrilineal herders from West Uzbekistan to East Kyrgyzstan. Bilineal means that there’s an even migration of men and women while patrilineal means there’s an uneven migration of women to their husband’s location. In total, their sample size represents 780 healthy adult men from 5 ethnic groups: Tajiks, Kyrgyz, Karakalpaks, Kazaks, and Turkmen. They conclude that the number of reproductive individuals is likely to be higher for women among patrilineal populations.

Both these studies show that the organization and structure of patrilineal populations is the likely cause of the observed genetic patterns, where men tend to father children with more females than females do with males despite institutionalized monogamy.

Michael F. Hammer, Fernando L. Mendez, Murray P. Cox, August E. Woerner, Jeffrey D. Wall, Dmitri A. Petrov (2008). Sex-Biased Evolutionary Forces Shape Genomic Patterns of Human Diversity PLoS Genetics, 4 (9) DOI: 10.1371/journal.pgen.1000202

Laure Ségurel, Begoña Martínez-Cruz, Lluis Quintana-Murci, Patricia Balaresque, Myriam Georges, Tatiana Hegay, Almaz Aldashev, Firuza Nasyrova, Mark A. Jobling, Evelyne Heyer, Renaud Vitalis, Molly Przeworski (2008). Sex-Specific Genetic Structure and Social Organization in Central Asia: Insights from a Multi-Locus Study PLoS Genetics, 4 (9) DOI: 10.1371/journal.pgen.1000200

Mark Stoneking is a population geneticist at the Max Planck anthropology powerhouse. He uses genetics to study the origin, relationships, structure and migration patterns of human populations. He’s written up a review of the origins of humans in the journal EMBO Reports under the title, “Human origins. The molecular perspective.” I’ve tried to get access to it, but my library doesn’t have a subscription and I’m not really willing to dish out $18 to buy the 4 page paper. I rather buy a book with that sort of cash.

But Dienekes seems to have read it and he’s found the illustration of the different models of human evolution interesting. I’ve decided to share the image with you. Dienekes also discusses Stoneking’s understanding of Carleton Coon, the physical anthropologist who wrote “The Origin of Races” in 1962, which was thwarted by the big boys of the time, Sherwood Washburn and Ashley Montagu and further discredited with Lewontin‘s 1972 conclusion that there’s more differences within groups than between.

We’ve come a long way since then. With DNA chips able to detect many SNPs and with the lower costs of genetic screening, we can assess the genetic diversity of more people. Datasets are growing and patterns are emerging. We’re now able to isolate distinguishable genetic differences in places like Europe. In the past, I’ve argued that with these higher resolution and more ubiqutious technologies available, we shouldn’t be falling back on Lewontin’s Fallacy anymore. Saying a biological component to race doesn’t exist just doesn’t cut it anymore.