June 12, 2012

ESHG 2012 abstracts

I had posted some interesting titles in April, and it is finally time for us, mere mortals, to have access to the abstracts. A few of the abstracts:


Title: P10.39 - Analysis of mitochondrial DNA haplotypes of old human populations from the Bronze and Iron Age from Romania

Keywords: mitochondrial DNA; haplotypes; Bronze and Age populations

Authors: A. Rodewald1, G. Cardos2, C. Tesio3;
Abstract: Our genetic study was focused on old human populations from the Bronze and Iron Ages from Romania in order to analysed their genetic variation and their genetic kinship al mitochondrial DNA(mtDNA)level with today´s Romanian populations and other modern European populations. The ancient DNA(aDNA)was extracted from skeletal remains of 50 individuals from the Bronze and Iron Age by a phenol-chloroform DNA extraction method.MtDNA HVR I and HVR II regions were amplified by PCR and sequenced by the dideoxy chain terminator method.The aDNA data were analysed in comparison with corresponding mtDNA data of modern Romanian people and other 11 European populations.The ancient mtDNA haplotypes were framed into 12 haplogroups. The most frequent mtDNA haplotype identified in the old individual sample from Romania was the CRS-like, and the most frequent haplogroup was H. Significant differences in haplogroup frequencies between the old people and modern Romanians were found. Low values of internal standard genetic diversity indices suggested reduced genetic variability within old human populations from the Bronze and Iron Age from Romania, in contrast to all modern European populations and also modern Romanians, which showed higher mitochondrial haplogroup diversity values. This fact might be the result of social and cultural local organization in small tribes, partially reproductively isolated. Concerning the genetic relationships at mitochondrial level, old human populations from Romania have shown closer genetic relationship to Turks of Thracian origin,while modern Romanians were closer to modern Bulgarian, Italian, Greek and Spanish populations.



The relationship between prehistoric Romanian mtDNA and modern Thracian Turkish one is not very surprising, given that the latter are basically descended from populations of indigenous origin that converted to Islam during the Ottoman era. It will be worthwhile to see exactly which locations/communities were used to sample from. It is awesome that we are beginning to get ancient DNA data from southeastern Europe.



Title: P10.40 - Complete mitochondrial DNA diversity in Iranians
Keywords: Mitochondrial DNA; population; Iran
Authors: A. Bahmanimehr1, M. Derenko2, B. Malyarchuk2, G. Denisova2, M. Perkova2, S. Farjadian3, L. Yepiskoposyan1;
Abstract: The complete sequencing of mitochondrial DNA has contributed a great deal to the understanding of the timing and direction of human dispersals around the world. To elucidate the early stages of human colonization process outside of Africa and to investigate the demographic history of human populations from the Middle East we have completely sequenced the mtDNAs of 275 Iranians represented by Persians (N=105), Mazandaranians (N=4), Azerbaijanians (N=22), Kurds (N=5), Lurs (N=5), Armenians (N=10), Bakhtiarians (N=2), Gilakis (N=2), Indians (N=1), Turkmens (N=10), and Qashqais (N=109). Overall diversity is very high, with 252 different sequences falling into 75 major haplogroups within macrohaplogroups L, N and M. The majority of Iranian mtDNAs (90.9%) belongs to Western Eurasian component composed of haplogroups N1, N2, X, R2’JT, U, and R0, though the impact of African (L2a, L3d, L3f), Southern Asian (R8, M4, M5, M18, M42), and Eastern Eurasian (A4, B4, C4, C5, D4, F1, G2a) lineages is also perceptible being found at frequencies of 1.5%, 2.5%, and 5.1%, respectively. Results of molecular dating of Iranian mtDNA lineages show that macrohaplogroup N and its haplogroups N1, R, U, R2’JT coalesce to the time of 45-60 kya, marking the first stages of modern humans movement out of Africa. The ancient ancestry of Iranian gene pool is also confirmed by revealing of the unique N23 lineage survived both in Persians and Qashqais, albeit at low frequencies. This study was supported by Russian Foundation for Basic Research (11-04-00620) and by Far-East Branch of the Russian Academy of Sciences (12-III-A-06-101).

Title: P10.58 - A Y -chromosome portrait of modern Bulgarians as viewed from different spatiotemporal aspects
Keywords: Y chromosome; Bulgarians; haplogroup
Authors: S. Karachanak1,2, V. Grugni2, D. Nesheva1, N. Al-Zahery2, V. Battaglia2, C. Nici2, V. Carossa2, Y. Yordanov3, A. Torroni2, A. S. Galabov4, O. Semino2,5, D. Toncheva1,5;
Abstract: To address the structure and evolution of the Bulgarian paternal gene pool, we have examined the Y chromosome variation in 809 Bulgarian males. The analysis was performed by high-resolution genotyping of biallelic markers and by analyzing the STR variation within certain haplogroups. The biallelic markers were analyzed by PCR/RFLP and PCR/DHPLC assay. Seventeen fast-evolving Y-STRs were amplified using the multiplex AmpFlSTR Yfiler PCR Amplification Kit (Applied Biosystems) and were read on ABI 310 genetic analyzer with GeneMapper software.We found that the Bulgarian Y chromosome gene pool is primarily contained within haplogroups common in Europe and surrounding areas. Furthermore, when patrilineal relationships are visualized in a broader context by principal component analysis, Bulgarians are located among European populations. The analysis of molecular variance shows that the genetic variation within the country is structured among Western, Central and Eastern Bulgaria, rather than among the Black Sea coast, the Danubian Plane, Thrace and the Southwest mountainous region; which indicates that the Balkan Mountains have been permeable to human movements.
Y-STR variation ages and median joining networks of haplogroups E-V13, J-M241, R-M458, R-L23 and I-M423 were calculated together with data from other populations. For this purpose, the analyses of STR variation within haplogroups were based on 8 STR loci, with the exception of haplogroup R-M458, for which the STR profiles were further reduced to 7 loci. In general, the Y-STR data reveal that different prehistoric and historic events have left detectable traces in the Bulgarian Y chromosome gene pool.

This is quite welcome, as Bulgaria has been a bit of a "black hole" in population genetics. Finally having an authoritative sample on its Y-chromosome composition is extremely important.


Title: P10.45 - Population isolates from Greece offer potential for powerful disease gene mapping: the HELIC-Pomak and MANOLIS studies
Keywords: population isolate; genetic association; rare variants
Authors: I. Tachmazidou1, A. Farmaki2, L. Southam1, K. Palin1, N. W. Rayner1,3, E. Daoutidou2, I. Ntalla2, K. Panoutsopoulou1, G. Dedousis2, E. Zeggini1;
Abstract: The study of low-frequency and rare variants can be empowered by focusing on isolated populations, in which rare variants may have increased in frequency and linkage disequilibrium tends to be extended. Sequencing is efficient in isolates, because variants are shared in extended haplotype contexts, supporting accurate imputation. Here we assess sample sets collected from two Greek populations: the Pomak villages are a set of religiously-isolated mountainous villages in the North (population size 11,000); Anogia is a mountainous village on Crete, with high levels of longevity (population size 4,000). 747 and 1118 individuals respectively were typed on the Illumina OmniExpress platform. We calculated genome-wide IBS statistics to assess the degree of relatedness and compared it with the general Greek population (707 samples with OmniExpress data, TEENAGE study). We additionally calculated the proportion of individuals with at least one “surrogate parent” as a means for accurate long-range haplotype phasing and imputation, as proposed by Kong et al, Nature Genetics 2008. We find 1-1.4% of individual pairs with pi-hat>0.05, and ~0.4% with pi-hat>0.1 in the isolates compared to 0% in the general Greek population. We also find that ~80%-82% of subjects have at least one surrogate parent in the isolates, compared to ~1% in the outbred Greek population. We have established the HELIC-Pomak and MANOLIS cohorts as genetic isolates and are currently whole-genome sequencing 250 individuals to enable imputation and subsequent association testing. This approach has the potential to identify novel robust associations with disease-related complex traits.
It will be wonderful if some of these population samples become publicly available.


13 comments:

Rob said...

RE "Analysis of mitochondrial DNA haplotypes of old human populations from the Bronze and Iron Age from Romania"

Their conclusion about apparent similarity of their BA & IA samples to modern Thracian Turks is not necessarily generalisable to all ancient Thracians; but merely for their sample data.

Rob said...

"A Y -chromosome portrait of modern Bulgarians as viewed from different spatiotemporal aspects"

Excellent. We need more studies like this. Ie looking at >>within<< country regional differences. Aggregating Y Haplogroups averages of countries can hinder rather than help efforts in constructing proper molecular anthropological conclusions.

To stick with the Balkans, we should look at north/ south Serbia, north / south Greece, east (of Struma) / west of Struma northern Macedonia, etc

Dienekes said...

Their conclusion about apparent similarity of their BA & IA samples to modern Thracian Turks is not necessarily generalisable to all ancient Thracians; but merely for their sample data.

That is a meaningless statement, that only someone ignorant of basic statistics would make.

Rob said...

"That is a meaningless statement, that only someone ignorant of basic statistics would make"

How so ? All I meant was that we cannot really be too confident that those ancient Thracians were "more" like modern Turks from Thrace than say , modern Romanians, from a mere 50 or so samples. (whislt their effort is most laudable, you'd need several hundreds of samples to have adequate power in their analysis, I thought that what's basic statistics is)

Dienekes said...

you'd need several hundreds of samples to have adequate power in their analysis

That is baseless speculation. There is no a priori sample size with "adequate power". For example, one could determine that Romanians and Nigerians have different mtDNA gene pools with a sample size of 5-10, whereas one would need a much bigger sample size to detect differences between two Romanian cities.

Also, the idea that "Their conclusion about apparent similarity of their BA & IA samples to modern Thracian Turks is not necessarily generalisable to all ancient Thracians; but merely for their sample data." shows a lack of understanding of what sampling is; the collection of samples from a population puts probabilistic contraints on the density of the population distribution, and hence ALWAYS generalizes to the non-sampled individuals.

Onur Dincer said...

from a mere 50 or so samples

50 samples is a high number for an ancient mtDNA sampling of such a restricted region from such a restricted period. Of course more is better, but 50 mtDNA samples is not a low number, especially for such an ancient population. These ancient mtDNA samples certainly have statistical value to make strong generalizations about population relationships on the maternal side.

genefan said...

All tries to position autosomal genetic results of Bulgarians and Romanians on the map show that they don't correspond to the current national area.
Bulgarians show more to the West, while Romanians to South West, proving there was some population movement in more recent times.
Many Turks on turn moved from North East Bulgaria to the area around Edirne and Istanbul.
So nobody can say who is more autochthonous at the current place of residence.
http://spittoon.23andme.com/wp-content/uploads/2011/09/Genes-mirror-geography-for-Europeans.jpg

Rob said...

Fair enough; anyhow; great study ! I suppose it'll be a year till full article is published

pconroy said...

The area of modern Ramania was also home to the Chernyakhov Culture - which was a blending of Goth, Dacian, Sarmatian

http://en.wikipedia.org/wiki/Chernyakhov_culture

Onur Dincer said...

In what is now Romania, the dominant group all through the Iron Age and probably also the Late Bronze Age if not earlier was Thracians (including Dacians and Getae). They gradually Romanized during the Roman rule. The Slavic migrations of the Middle Ages probably led to some changes in the genetic makeup of the area. As maternal lineages better reflect autosomal genetic variation than paternal lineages, the most plausible conclusion from the results of this study is that: the Bronze and Iron Age people of what is now Romania are genetically closest to Thracian or Balkan Turks among modern populations, and that is because that Balkan Turks are mostly from the southeastern parts of the Balkans and that the southeastern parts of the Balkans are probably among the parts of the Balkans that were least affected by the Slavic migrations and have preserved the pre-Slavic genetic structure the most.

Unknown said...

"Certainly if prehistoric Romanians are most like living Thracian Turks, this suggests that there was a population movement in historical times into the Balkans, probably related to the Slavs."

I presume you mean dilution of an ancient Thrace-like population of Romania by incoming Slavs?

It could also be taken as evidence of a movement of Slavs in to Thrace.

This is after all evidence that modern Thrace is like ancient Romania. Not that modern Romania is like ancient Thrace.

terryt said...

"The ancient ancestry of Iranian gene pool is also confirmed by revealing of the unique N23 lineage survived both in Persians and Qashqais, albeit at low frequencies"

I presume that is a mispront. Do they mean N3, fairly recently discovered in Iran? N23 is certainly not listed in the April version of Phylotree.

terryt said...

Another extremely interesting aspect of that paper:

"Results of molecular dating of Iranian mtDNA lineages show that macrohaplogroup N and its haplogroups N1, R, U, R2’JT coalesce to the time of 45-60 kya, marking the first stages of modern humans movement out of Africa".

That is comfortingly close to the dates Behar came up with in the recent paper Dienekes blogged here:

http://dienekes.blogspot.co.nz/2012/04/copernican-reassessment-of-human.html

Perhaps the dates in the current paper are based on those ones. That paper has N first spreading at 58,859.9 ya., N1'5 at 56,547 ya., and N1 at 51,642.7 ya. R spread at 56,523.5 ya., and R2'JT at 53,771.5 ya. and U at 46,531 ya. Those dates fit completely within the 45-60 kya date claimed in the current paper.

But by 53,771.5 ya. R2'JT began its spread P had already begun spreading from Australia to new Guinea and the Philippines (54,801.3 ya.) and R31 had begun spreading round India (54,895 ya.). As I said, very interesting.