A common founder for the 35delG mutation

Am J Med Genet A. 2008 Oct 16. [Epub ahead of print]

Strong linkage disequilibrium for the frequent GJB2 35delG mutation in the Greek population.

Kokotas H, Van Laer L, Grigoriadou M, Iliadou V, Economides J, Pomoni S, Pampanos A, Eleftheriades N, Ferekidou E, Korres S, Giannoulia-Karantana A, Van Camp G, Petersen MB.

Department of Genetics, Institute of Child Health, Athens, Greece.

Approximately one in 1,000 children is affected by severe or profound hearing loss at birth or during early childhood (prelingual deafness). Up to 40% of congenital, autosomal recessive, severe to profound hearing impairment cases result from mutations in a single gene, GJB2, that encodes the connexin 26 protein. One specific mutation in this gene, 35delG, accounts for the majority of GJB2 mutations detected in Caucasian populations. Some previous studies have assumed that the high frequency of the 35delG mutation reflects the presence of a mutational hot spot, while other studies support the theory of a common founder. Greece is among the countries with the highest carrier frequency of the 35delG mutation (3.5%), and a recent study raised the hypothesis of the origin of this mutation in ancient Greece. We genotyped 60 Greek deafness patients homozygous for the 35delG mutation for six single nucleotide polymorphisms (SNPs) and two microsatellite markers inside or flanking the GJB2 gene. The allele distribution in the patients was compared to 60 Greek normal hearing controls. A strong linkage disequilibrium was found between the 35delG mutation and markers inside or flanking the GJB2 gene. Furthermore, we found a common haplotype with a previous study, suggesting a common founder for the 35delG mutation.

Link

ASHG 2008 abstracts

Just a sample of abstracts that I found interesting from the upcoming meeting of the American Society of Human Genetics.

Strong linkage disequilibrium for the frequent GJB2 35delG mutation in the Greek population.

Up to forty percent of autosomal recessive, congenital, severe to profound hearing impairment cases result from mutations in the GJB2 gene. The 35delG mutation accounts for the majority of mutations detected in Caucasian populations and represents one of the most frequent disease mutations identified so far. Some previous studies have assumed that the high frequency of the 35delG mutation reflects the presence of a mutational hot spot, whilst other studies support the theory of a common founder. Greece is amongst the countries presenting the highest frequency of the 35delG mutation (3.5%), and a recent study raised the hypothesis of the origin of this mutation in ancient Greece. We genotyped 60 Greek deafness patients homozygous for the 35delG mutation for six single nucleotide polymorphisms (SNPs) and two microsatellite markers, mapping within or flanking the GJB2 gene, as compared to 60 Greek hearing controls. A strong linkage disequilibrium was found between the 35delG mutation and the DNA markers at distances of 34 kb on the centromeric and 90 kb on the telomeric side of the gene, respectively. A comparison of the present findings with those of a previous study from Belgium, UK and USA, demonstrated a common haplotype reflecting the common founder. Our study supports the hypothesis of a founder effect and we further propose that ethnic groups of Greek ancestry could have propagated the 35delG mutation, as evidenced by historical data beginning from the 15th century BC.

Detection of population substructure among Jews and a north/south gradient within Ashkenazi Jews using 32 STR markers.

Understanding and detecting population substructure are critical issues. Using 32 autosomal STR markers and the program STRUCTURE we demonstrated differentiation between Ashkenazi (AJ) (N=135) and Sephardic (SJ) (N=226) Jewish populations in the form of Northern and Southern European genetic components (AJ north 73%, south 22%, SJ north 32%, south 61%) and a significant relationship between latitude of grandparental country of origin (GCO) and percent north/south genetic component in AJ. Notably, we revealed substructure among Jews (and among European Americans (EA)) using a small STR panel, only when additional samples representing major continental populations (African American, EA, Asian) were included in analyses. Further, negative RIS (-0.035) indicates recent admixture in individuals with both SJ and AJ parents (N=38). RIS is a measure of inbreeding adapted from FIS for STR markers. Negative RIS indicates allelic variation within individuals greater than expected under random mating, i.e., excess heterozygosity due to outbreeding. Although geographic patterns are seen in the average north/south percent assignment values between groups as defined by AJ or SJ, grandparental world region of origin, or GCO, within each group there is high variability among individual assignment values. Thus, even based on data from a small marker set, AJ is not a homogeneous population. The north/south gradient in AJ may be a reflection of the pre-existing north/south gradient in European host populations (recently shown in other studies using large numbers of SNPs) with which Jews admixed slowly. We also demonstrate the utility of including purported parental populations when attempting to detect population substructure within closely related populations.

Mutation meltdown of mitochondrial DNA and Neanderthal extinction.

There is emerging evidence that mitochondrial DNA (mtDNA) plays and integral role in the evolution of the human species. Although contentious, recent phylogenetic studies of modern humans implicate genetic variation of mitochondrial DNA (mtDNA) as a major factor underpinning the climatic adaptation of across the globe. Greater sequence diversity in the MTATP6 gene in arctic populations led to the idea that specific mtDNA polymorphisms cause subtle uncoupling of the respiratory chain, with the subsequent generation of additional heat being adaptive in northern climes. Our knowledge of mtDNA and its affect on adaptability may help us to understand how modern humans have survived their early ancestors. Here, we characterise the mtDNA of one of these extinct hominids. Neanderthals are the closest hominid relatives of modern humans, who up until 30,000 years ago coexisted in Europe and western Asia. Recently, over 1Mb of DNA was successfully extracted and characterised from the Vi-80 Neanderthal fossil. We reanalysed 2,705 base pairs of mtDNA in order to examine the hypothesis that mitochondrial dysfunction contributed to the Neanderthals demise. We identified thirty-two nucleotide differences from the modern human mtDNA reference sequence and by treating the Vi-80 as a diagnostic sample leads us to the conclusion that sequence variants that are highly likely to be artifacts, and a large proportion of the remaining mutations could be due to nuclear pseudogene amplification. We did identify a potentially deleterious variation; however more study may be needed to ascertain the effect of mitochondrial dysfunction on Neanderthal survival.

Early Siberian Maternal Lineages in the Tubalar of Northeastern Altai Inferred from High-Resolution Mitochondrial DNA Analysis

At the hight of the last glaciation (~18 kya) Siberians were confined to the southern strongholds, which were areas of continuous occupation, and where immediate ancestors of the Uralic, Kettic and Altaian language groups differentiated. To better understand the evolutionary relationships between the earlier and contemporary Siberians, we focused on the northern Altaic prehistory preserved in the mtDNA diversity of the Tubalar, until recently representing a typical hunting-gathering population. The present study includes 139 Tubalar. All mtDNAs were subjected to high-resolution SNP analysis, followed by complete sequencing of selected mtDNA samples. We showed that the core of the Tubalar genetic makeup proved to be a mixture of west (H8, U4b, U5a1, and X2e) and east Eurasian (A and B1) haplogroups derived from macrohaplogroup N, and Siberian derivatives of the macrohaplogroup M identifiable by subhaplogroup-specific mutations. For example, among the 36 Tubalar mtDNA samples that belong to haplogroup D, 10 (28%) harbored diagnostic markers of the subhaplogroup D3a2a shared with the Chukchi and Eskimos. This finding verified at the complete sequence level we attributed to ancient link between early Siberians, who underwent pronounced differentiation in the Altai-Sayan region, and some of the Eskimo tribes. A comparison of the mtDNA data generated through the course of this study with published complete sequences has contributed essentially to parsimonious phylogenetic structure of mtDNA evolution in west Siberia. Specifically, northeastern Altai appears to be a good candidate for the ancestral homeland of the haplogroup U4b, which is apparently ancient European. For some haplogroups, such as X2e, the relatively recent arrival to the Altai region is more likely.

Sex-specific gene flow between Pygmy and non-Pygmy populations

Cultural traditions and preferences may drive sex-specific gene flow among human populations. We have examined sex-specific gene flow between Mbuti Pygmies, a hunter-gather population, and surrounding agriculturist groups, the Alur, Hema, and Nande, which all reside in Central Africa. We used 18 lineage-defining Y chromosome SNPs and HVS1 mitochondrial DNA sequence information to examine patterns of gene flow among these groups. Mbuti Pygmy males have more diverse Y chromosome lineages (Mbuti Pygmy [n = 28]: = 0.229; Alur [n = 10]: 0.193; Hema [n = 18]: 0.178; Nande [n = 15]: 0.090) and slightly less mtDNA diversity than neighboring groups (0.020, 0.023, 0.025, 0.022 in Mbuti Pygmy, Alur, Hema, and Nande groups, respectively). The majority of Mbuti Pygmy males have a Y haplotype characteristic of Mbuti Pygmies (B2b); however, more than 30% of Pygmy males exhibit Y haplotypes associated with Bantu-speaking agricultural populations (E3a lineage). Conversely, no agriculturist males exhibit Y haplogroups associated with Mbuti Pygmy populations but instead have derived Y haplogroups characteristic of Bantu agriculturalists (E2, E3a). Pairwise FST was calculated among all populations using Y haplogroup frequency and HVS1 mtDNA sequence data. YDNA and mtDNA FST values between Mbuti Pygmy and non-Pygmy groups (Alur, Hema, and Nande) were 0.278, 0.355, and 0.217 (for YDNA) and 0.088, 0.239 and 0.217 (for mtDNA), respectively. A Mantel test between pairwise FST matrices showed no significant correlation ((r = 0.27; p 0.35), which indicates that patterns of genetic differentiation differ between Y chromosome SNPs and mtDNA sequence patterns. These results also suggest no emigration of Mbuti Pygmy Y chromosomes into surrounding groups but immigration of non-Mbuti Pygmy Y chromosomes into the Mbuti Pygmy population.

Population Structure in Mongolia from a Mitochondrial DNA Perspective.

Mongolia has experienced a complex series of demographic movements over the past 10-20 millennia that have shaped the patterns of its modern human genetic variation. However, modern populations in Mongolia have not been extensively studied for DNA diversity, nor has the genetic contribution of Mongolians to the gene pools of contemporary populations in Southeast Asia and Oceania been fully resolved. Archaeological evidence from as early as the late Neolithic suggests the presence of both West and East Eurasian cultures in this region. Later demographic movements involving the emergence of the Mongolian and later Manchu Empires have further convoluted Mongolias population structure. To clarify the complex population history of Mongolia, we analyzed variation in the mtDNAs of 190 individuals from several Mongolian ethnic groups, including the Uriankhai, Zakhchin, Derbet, Khoton and Khalkha. We screened all samples for phylogenetically informative coding region SNPs and sequenced HVSI to assess control region variation in them. Our data suggest that the mtDNA diversity present in our population is consistent with the general pattern of variation observed in East Asia, with the most frequent haplogroups being C, D and G. Haplogroup variation in Mongolian ethnic groups reveals considerable maternal diversity with a predominance of basal M types. Interestingly, the Mongolians also possessed West Eurasian haplogroups, such as H, J and K, which are not commonly observed in East Asia, even at low frequencies. The main ethnic group in Mongolia, the Khalkha, was highly variable with respect to mtDNA haplotypes in comparison with the other ethnic groups, and clearly distinct from the Khoton and Zakhchin, as evidenced by distance measures. Overall, these data provide insights into the origins and affinities of these populations, their relationships with East Asian groups and neighboring Turkic speaking groups, including indigenous Altaians, and their possible role in the peopling of the Americas.

Allocation of YSTR Microvariant Alleles to Y-Chromosome Binary Haplogroups.

Y-chromosome short tandem repeat (YSTR) loci are used extensively in studies of population substructure, temporality of population dynamics, and forensic identification. The occurrence of non-consensus YSTR alleles, such as unusually short alleles or partial insertion/deletion events (microvariants), have been used successfully as indicators of common ancestry among YSTR haplotypes, exposing further levels of phylogenetic substructure with restricted geographic distributions. However, the high variability of STR loci can potentially lead to false associations due to homoplasy (ie, recurrent mutation). Thus, YSTR haplotypes are best interpreted within the context of the binary marker defined Y-chromosome phylogeny. To identify YSTR microvariant alleles potentially useful for elucidating further phylogenetic substructure within binary haplogroups, we have assessed the haplogroup affiliation of microvariant alleles found at informative frequencies in public YSTR databases for the following YSTR loci: DYS385, DYS392, DYS441, DYS446, DYS447, DYS449 and DYS464. We report haplogroup affiliations for each variant allele and geographic origins of representative samples.

L1c2a, the (African) Haplogroup With The Longest Mitochondrial Genome!

Haplotypes derived from the maternally-inherited mitochondrial DNA (mtDNA) control region are often employed as a first step in determining phylogenetic-relevant samples that could be selected for additional coding region testing. Using the currently defined world mtDNA haplogroup tree, researchers can assign these haplotypes to specific branches, paying particular attention to novel mutations that could assist in identifying new subclades. During a recent survey of the nearly 58000 mtDNA control region haplotypes currently present in the publicly accessible Sorenson Molecular Genealogy Foundation database, we observed a small number of mtDNAs (n=16) characterized by the presence of unusually long insertions of up to 200 bases. A small subset of these particularly long mtDNA haplotypes shared an identical insertion of 15 bases. Genealogical analysis combined with haplogroup prediction confirmed that these haplotypes shared a common African origin. Additionally, based on the pedigree data gathered, we determine the donors were not closely related. Moreover, through the analysis of complete mtDNA sequences, we conclude that the newly defined haplogroup is most likely of recent origin. As reported in this study, insertions of more than 10 bps are quite rare in the general population and in the published literature, thus providing an interesting case work in population and possibly future disease studies.

Mitochondrial DNA footprints in modern Mongolia.

Although Mongolia is one of the most sparsely populated countries in the world, it is located at a pivotal crossroad between the four corners of Asia (including the well-known Silk Road) and has been characterized throughout history by events that greatly added to its current cultural and ethnic diversity. Among these, perhaps one of the most significant happening was the ambitious expansion strategy employed by Mongolias most prominent personality, Genghis Khan, whose empire eventually stretched across all of modern-day China, a portion of modern Russia, Southern Asia, Eastern Europe and the Middle East. In 2007, through a well-planned collection effort, researchers at the Sorenson Molecular Genealogy Foundation and the National University of Mongolia were able to gather over 3,000 DNA samples, informed consents, and genealogical data throughout the country of Mongolia, including samples from 21 distinct tribal or ethnic populations. All the samples were sequenced for the three hypervariable segments of the mitochondrial DNA (mtDNA) control region to assess the genetic composition of modern Mongolia. The most common mtDNA haplotypes are typical of haplogroup C, which is frequent throughout Eastern Asia. However, nearly 40% of the observed mtDNA lineages are of Western Eurasian origin, including a significant frequency (~7%) of haplogroup H - the most common in Europe. The high prevalence of Western Eurasian lineages could be a remnant from Genghis Khans conquering efforts, trade and cultural exchanges along the Silk Route. To assess the extent of recent gene flow that could account for the elevated levels of Eurasian haplogroups within Mongolian populations, we have examined genealogical data of samples representative of Western Eurasian haplogroups.

Y chromosome microsatellite haplotypes in the Hutterite founders.

The current population of >12,000 Schmiedeleut Hutterites are descendants of 38 male founders who were born between 1700 and 1830 in Europe. Only 12 of these founders, each with a unique surname, have living male descendants related through male-only lineages. DNA samples were available in our laboratory for 75 male descendants of 11 of the 12 founders, accounting for 673 independent paternal meioses. We genotyped 9 microsatellite loci, which included a mean of 6.8 (range 2-23) males per lineage to evaluate potential relationships between the founders. Fourteen different haplotypes were identified, with an average of 3.5 (range 1-8) pairwise differences between haplotypes. All descendants within each of 9 lineages had identical Y haplotypes. Descendents of two of these lineages, 2 and 10, had the same haplotype despite different surnames, suggesting possible relatedness between the founders of these two lineages. Descendants of two lineages, 6 and 11, each carried three distinct haplotypes. Within each of these lineages the haplotypes differed from the ancestral haplotype by one repeat size at two loci. Additional male descendants in lineages 6 and 11 were then genotyped for the discrepant microsatellites, confirming the presence of three Y haplotypes each in lineages 6 and 11. The one mutation arose at each of four loci: DYS388, DYS389II, DYS390, DYS393. Three mutations were gains of one repeat; it was not possible to determine if the fourth mutation was a gain or loss of one repeat. The ancestral haplotypes in these two lineages are identical at four microsatellite loci; the alleles at the other five loci differ by one repeat size. The average mutation rate at these 9 loci was 0.00066 (95% CI 0.00015-0.0013), similar to other estimates. These data suggest that the founders of lineages 2 and 10 may have been related through paternal lines and that surnames do not strictly correspond to unique Y chromosomes. Moreover, certain ancestral haplotypes (i.e., those in lineages 6 and 11) may be more prone to mutation. Supported by NIH grants HD21244 and HL085197.

Genetic History of human populations of East African inferred from mtDNA and Y chromosome analyses.

Evidence from genetic, paleobiological, and archaeological studies suggest that Africa, especially East Africa, is most likely to be the cradle of the modern human species. Despite this fact, very little is currently known about genetic diversity in African populations in general, and East African populations in particular. Genetic data demonstrate that the patterns of genetic variation in East African populations are complex. All four major language families spoken in Africa (Afro-Asiatic, Nilo-Saharan, Niger-Kordofanian, and Khoisan) are found in the region. As part of a large study of population genetic diversity of East and Northeast Africa, we examined Y chromosome genetic diversity (to ascertain paternal lineages) as well as mitochondrial genetic diversity (to ascertain maternal lineages) in 1200 - 1500 individuals from ~ 40 Tanzanian, Sudanese, and Kenyan populations. For the Y chromosome analysis, we genotyped 60 UEPs (analyzed in a hierarchical manner to construct haplotypes) in a total of ~1500 male individuals. In order to infer ages of lineages and migration patterns, we further genotyped the individuals for 16 Y chromosome microsatellites. For the mtDNA analysis, we sequenced the mitochondrial D-loop in a total of 1200 individuals from the same populations, and for 200 individuals, we did complete mitochondrial genome sequencing. We compare our results with published results of studies from other parts of Africa and the Middle East. Our results indicate that East African populations have some of the most ancestral Y chromosome and mtDNA lineages in Africa, suggesting that they may have been an ancient source of dispersion throughout Africa. Additionally, we find evidence for ancient geneflow between East Africa and the Middle East. We also ascertained the effect of the Bantu-expansion and signature of recent migration of Cushitic-speaking groups originating from Ethiopia on peopling of East Africa.

Analysis of mtDNA and Y-chromosome haplogroups in Mexican Mestizos and Amerindian groups.

The Mexican population is mainly conformed by Mestizos, individuals with a genetic background consisting of Amerindian, European and African contributions. Genetic heterogeneity in Mexicans results from a complex demographic history that started with the peopling of North and Central America about 15,000 yrs ago, including the settlement of at least 60 different indigenous groups in Mexico, regional differences in admixture dynamics after colonization by Spaniards in the XVI century, epidemics and migration. Y chromosome-specific and mitcohondrial (mt) DNA polymorphisms are useful to help understand the genetic structure and history of human populations, due to their uniparental inheritance and lack of recombination. In order to refine the portrait of genetic variability derived from the Mexican Genome Diversity Project, we are characterizing maternal and paternal lineages participating in admixture. For this we included genotypic data from 163 mt SNPs and 123 Y chromosome SNPs present in the Illumina Human1M chip of 450 individuals, 300 mestizos from six states located in different regions: Northern, Central and Southern; and 150 individuals from different Amerindian groups (Tepehuanes, Zapotecos and Mayas). With this information, we are measuring genetic diversity using Fst and AMOVA analysis. Admixture analysis includes average and individual ancestral contribution estimates using autosomal SNPs. Initial results show that in our Mestizo sample, 88% of the mt haplogroups are Amerindian (A, B, C or D), and the rest includes European and African lineages. We have identified differences in proportions of each haplogroup in both Mestizos and Amerindians. Knowledege about the distribution of mt and Y-chromosome haplogroups in Mexican Mestizos and Amerindian groups, will generate valuable information to better understand genetic relationships between Mexicans and other Latin American populations. In addition, it may contribute to strengthen analysis in association studies of common complex diseases.

The origin of Native Americans from a mitochondrial DNA viewpoint.

America, the last continent to be colonized by modern humans, is characterized by an extraordinary linguistic and cultural diversity. Until recently, it was generally believed that starting around 13,500 years ago, the first Paleo-Indians arrived from Beringia, passing through an interior ice-free corridor in western North America, and spread rapidly all the way to Tierra del Fuego. Today, we realize that the peopling of the Americas involved a much more complex process. As for the maternally transmitted mitochondrial DNA (mtDNA), it has been clear since the early nineties that Native Americans could be traced back to four major maternal lineages (haplogroups) of Asian affinity. These were initially named A, B, C and D, and are now termed A2, B2, C1 and D1. More than 95% of living Native Americans belong to these four haplogroups, which can be considered pan-American, because they are shared by North, Central and South American populations. Later, five additional maternal lineages were discovered and named X2a, D2, D3, C4c, and D4h3. These less common or rare haplogroups are restricted only to some Native American populations or geographic areas and bring the overall number of Native American mtDNA lineages to nine. Our comprehensive overview of the four pan-American branches of the mtDNA tree suggests a scenario with a human entry and spread into the Americas from Beringia about 20,000 years ago, and preliminary data raise the possibility that the uncommon five Native American haplogroups might have marked additional migratory events from Asia or Beringia. Overall, through a combined analysis of modern and ancient Native American mtDNA, we are making an effort for reconstructing the complex pre-Columbian history at both macro- and micro-geographic levels.

Identifying genes affecting normal variation in human facial features using admixed populations.

Seven selection-nominated candidate genes (COL11A1, LMNA, FGFR1, FGFR2, TRPS, BRAF, FLNA) known to be involved in Mendelian craniofacial dysmorphologies and to have high allele frequency differences between West African and European populations were tested for admixture linkage to normal facial feature traits. The sample consists of 254 subjects (n=131 African Americans, n=123 Brazilians) of West African and European genetic ancestry. Each individual was genotyped at 176 ancestry informative markers (AIMs), which allowed for proportional estimation of genetic ancestry from four parental populations and adjustments for admixture stratification.
3D images of faces were acquired using the 3dMDface imaging system. 3D coordinate data were collected from 22 landmarks placed on each image using the 3dMDPatient software. The 231 possible pairwise landmark distances were scaled to the geometric mean and then analyzed using Euclidean Distance Matrix Analysis.
We used both ANOVA and ADMIXMAP to control for admixture stratification and to test for associations between the 231 pairwise landmark distances and 183 AIMs, using sex, height and BMI as covariates. We used a four-population model (West African, European, East Asian, and Native American).
There is a strong concordance between the ANOVA and ADMIXMAP results. Many landmark distances, particularly on the mouth and nose, were significantly associated with genetic ancestry. Additionally, three of the candidate genes show no effects on pairwise landmark distances while four show distinct patterns of association. For example, FGFR2 is associated primarily with the length of the face. These results represent the first identification of the first genes affecting normal variation in facial features.

Ethnicity-Confirmed Genetic Structure in New Hampshire.

Genetic population structure is known to result from shared ancestry. Though there have been several studies of genetic structure within and among different geographic regions and ethnic groups, little is known of the genetic structure of highly admixed US populations or whether the structure is concordant with self-reported ancestry. In this study, 1529 single nucleotide polymorphisms (SNPs) from 864 healthy control individuals from New Hampshire were measured as part of a bladder cancer epidemiology study. The SNPs were from approximately 500 cancer susceptibility genes scattered throughout the genome. Of these, 960 Tag SNPs were used to cluster individuals using the Structure algorithm for between 2 and 5 subpopulations. Subtle genetic structure was found, suggesting the appropriate number of subpopulations to be either 4 or 5 (FSTs 4 populations: 0.0377, 0.0399, 0.0363, 0.0340; 5 populations: 0.0452, 0.0536, 0.0585, 0.0534, 0.0521). We coded the individuals self-reported ancestries in a genotype fashion (i.e. 0= not reporting that ancestry, 1= reporting part that ancestry, 2= reporting only that ancestry) and conducted a Spearmans rank correlation between each ancestry and the structure q value, which represents the proportion of an individual that originated from a certain genetic subpopulation. Those of Russian, Polish and Lithuanian ancestry most consistently clustered together. The ancestry results support either 4 or 5 subpopulations. In order to investigate linkage disequilibrium (LD), the complete set of SNPs from the 7 most densely genotyped genes were used to make haploview plots between the different groups. The results vary by gene, though for one gene in particular, GHR, the results are very different for 4 subpopulations. These results suggest that despite New Hampshires admixture and presumed homogeneity, there are 4 or 5 distinct genetic subgroups within the population that can be linked to self-reported ancestry and display differences in patterns of LD.

Inference of human demographic parameters using haplotype patterns from genome-wide SNP data.

Accurate inference of human demographic history from genetic data is essential for identification of single nucleotide polymorphism (SNP) association with disease and for inference of natural selection. Haplotype diversity and haplotype sharing carry additional demographic information to that obtainable from SNP frequency spectra, and so we propose a novel method using haplotype summary statistics to fit demographic models to genome-wide SNP data. We divide the genome into 0.25 cM windows and for each we tabulate the number of distinct haplotypes and the frequency of the most common haplotype. We summarize the data by the genome-wide joint distribution of these two statistics. Coalescent simulations are then used to evaluate whether different demographic models are compatible with the observed data. Application of our method to simulated data shows that our method can reliably infer parameters from complex demographic models (such as bottlenecks) and is relatively robust to the levels of SNP ascertainment bias found in many genome-wide datasets. We have applied our method to data collected by the International HapMap Consortium and find that a bottleneck model best fits the CEU population. We have also analyzed a large dataset consisting of Affymetrix 500k data from ~2,900 individuals with ancestry from Taiwan, Japan, India, Mexico and many European countries. Since this dataset includes ~2,300 European individuals, we are able to study haplotype patterns at a fine scale within Europe. Interestingly, we find that within Europe there is a south-to-north gradient with decreasing levels of haplotype diversity moving north, consistent with south to north migrations. We also find that the southwestern European sample has higher haplotype diversity than the southeastern European sample. Additionally, a higher proportion of haplotypes are shared between the southwestern European sample and the Yoruba sample than between southeastern European sample and the Yoruba sample. These two patterns are consistent with recent admixture across the Mediterranean from Northern Africa.

Genome wide analysis and heritability estimation of intelligence in the International Multi-centre ADHD Genetics (IMAGE) study.

Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder characterised by symptoms of inattention, hyperactivity and impulsivity. There is growing evidence of heterogeneity in its etiology, pathophysiology and clinical expression. One approach to resolving heterogeneity involves the identification of endophenotypes, intervening variables that might mediate pathways between specific genes and clinical phenotype. IQ is a candidate endophenotype for ADHD. Genome-wide linkage analyses of full scale IQ and IQ subscales were performed in the International Multi-centre ADHD Genetics (IMAGE) study including 1094 families with 1094 DSM-IV combined type ADHD probands and their 1441 siblings (unselected for ADHD status). IQ was measured using five subscales of the WISC-IIIR scale. The full scale prorated IQ score and the five subscales were used as quantitative traits for linkage analysis. 5,407 autosomal SNPs were used to run multipoint regression-based linkage analyses using MERLIN. The h2 estimates from the IQ subscales and the full IQ score ranged from 31% to 100%. Three suggestive linkage signals were found (LOD scores 2, p values 0.001) on chromosomes 7, 9 and 14 for three different subscales. Previously, two regions on chromosomes 7 and 14 were reported as being associated or linked to IQ. Our results, though only suggestive, suggest the presence of additional genetic variants contributing to the variance of IQ in ADHD.

ESHG 2008 abstracts

The European Society of Human Genetics conference is coming up, and there are some very interesting abstracts.

Note: The ESHG site has updated with a notice that the abstracts are embargoed until their presentation time. Therefore, I have decided to remove the body of this post until then, although I think it is a bit weird to embargo something that one places on the public web. In any case, you can find the abstracts easily by going to the site above. (June 1): post restored.

The peopling of North Asia: Y and X perspectives

V. A. Stepanov, V. Kharkov, I. Khitrinskaya, O. Medvedeva, M. Spiridonova, A. Marusin, V. Puzyrev; Institute for Medical Genetics, Tomsk, Russian Federation.

Presentation Number: P07.056

To reconstruct the origin and evolution of human populations in North Asia we investigated the genetic diversity in 50 population samples (about 2000 individuals totally) using Y and X chromosome lineages. Y-chromosomal haplotypes were constructed with unique event polymorphisms (UEP) and STR markers according to Y Chromosome consortium (YCC) classification. SNP markers in a single 60 kb linkage disequilibrium region of ZFX gene was used to trace the X chromosomal population history. The genetic diversity of Y haplogroups was quite high (0.70 - 0.95) in most populations except few very isolated groups. The proportion of inter-population differences in the total genetic variability measured by Fst statistics is 17% for binary haplogroups and 19% for YSTR. Multidimensional scaling and principal component analysis revealed four major components in North Asian Y gene pool, reflecting the presence of Paleoasiatic (Q), Proto-Uralic (N3, N2), Eastern Asian (O, C), and Western Eurasian (R1, I, J) lineages. X-chromosomal haplotypes in North Asia are less divers (gene diversity within populations 0.65 - 0.80) and less differentiated (Fst = 4%) compared to Y lineages. The population clustering by X and Y gives, to a first approximation, a similar picture, and matrixes of genetic distances between populations for X and Y haplotypes significantly correlates. The age of genetic diversity generation and time of population differentiation demonstrates the Upper Paleolithic origin of major Y and X lineages and post-glacial population expansions. This work is supported by RFBR grants ##06-04-48274 and 07-04-01629.

The following seems to be a very important study; in particular the notion that particular Y chromosome/mtDNA haplogroups may be associated with higher or lower fertility may have implications about their distribution.

UPDATE (May 21): I did a quick and dirty analysis of the Y-haplogroup and mtDNA-haplogroup data from Bosch et al. (2006) (Ann Hum Genet. 2006 Jul;70(Pt 4):459-87.), and there is a -0.43 correlation between Y-haplogroup I and mtDNA-haplogroup H and a +0.46 correlation between Y-haplogroup R1 and mtDNA-haplogroup H. While not significant (with only 10 populations), this is definitely in the right direction for a selection effect for/against specific Y-DNA/mtDNA combinations.

... on the other hand, another quick and dirty analysis of 23 populations from Rootsi's survey on Y-haplogroup I and mtDNA frequencies from AJHG Volume 80, Issue 4, April 2007, Pages 759-768 didn't turn up any correlation. Perhaps, someone can look at possible correlations between Y-chromosome and mtDNA haplogroups in Europe to see if anything interesting turns up.

Male infertility induced by mtDNA/Y unfavorable combination? An association study on human mitochondrial DNA

S. C. Gomes1, S. Fernandes2, R. Gonçalves1, A. T. Fernandes1, A. Barros3, H. Geada4, A. Brehm1; 1Human Genetics Laboratory, University of Madeira, Funchal, Portugal, 2Genetics Department, Faculty of Medicine, University of Porto, Porto, Portugal, 3Centre of Reproductive Genetics A Barros, Porto, Portugal, 4Faculty of Medicine, University of Lisbon, Lisboa, Portugal.

Presentation Number: P07.084

There is growing evidence that certain mtDNA haplogroups determine a genetic susceptibility to various disorders bringing out the interest in the possible role of mtDNA background on the phenotype expression of mitochondrial genetic disorders. An association between haplogroup T and asthenospermia has been reported and several sublineages of haplogroup U were associated with differences in sperm motility and vitality. The deletion of some DAZ copies gene in 10-15% of azoospermic and oligospermic patients has been reported but also present in fertile men belonging to certain Y-haplogroups. The findings of one study have rarely been replicated by studies in other populations and conflicting associations have been reported. Our focus in this case-control study is to investigate the existence of other influences, besides a weak mtDNA background, promoting male infertility. The occurrence of a specific mtDNA variant associated to a certain Y-chromosome haplogroup could represent a vital link that will compromise the sperm function and be responsible for male infertility. A group of 99 infertile men and other one composed by 90 subjects with proven fertility were selected and analysed. The frequency of the combination mtDNA-haplogroup H (especially with the CRS sequence) and Y-haplogroup R was higher in fertile than in infertile men seemingly to be favorable to fertility. On the other hand, a considerable number of infertile men belonging to mtDNA-haplogroup H (CRS) and to Y-haplogroup I, associated to a specific DAZ gene deletion pattern- 2+4d, suggests a non favorable combination to male fertility.

The Bayash Roma: phylogenetic dissection of Eurasian paternal genetic elements

I. Martinovic Klaric, M. Pericic Salihovic, L. Barac Lauc, B. Janicijevic; Institute for Anthropological Research, Zagreb, Croatia.

Presentation Number: P07.110

The Bayash consist of numerous and small Romani groups speaking different dialects of the Romanian language and living dispersedly in Croatia, Hungary, Bosnia and Herzegovina, Serbia, Romania, Bulgaria, and to the lesser extent in Macedonia, Greece, Ukraine, Slovakia and Slovenia. Larger Bayash groups migrated to Croatia most likely during the 19th century, after abolition of slavery in Romania. Molecular architecture and the origin of the Croatian Bayash paternal gene pool was addressed by analysing 151 Bayash Y chromosomes from two Croatian regions, 332 Y chromosomes from Romani populations across Europe, 814 Y-chromosomes from non-Romani host populations living in Southeastern, Southern and Eastern Europe as well as with 1680 Y-chromosomes from South Asian populations. The Bayash in Croatia represent one population of largely shared paternal genetic history characterized by substantial percentage (44%) of common H1-M82 and E3b1-M78 lineages. Relatively ancient expansion signals and limited diversity of Indian specific H1-M82 lineages imply descent from closely related paternal ancestors who could have been settled in the Indian subcontinent between 7th and 9th centuries AD. Minimal time divergence of the Bayash subpopulations is consistent with their putative migratory split within Romania towards Wallachia and Transilvania. Substantial percentage of E3b1 lineages and high associated microsatellite variance in the Bayash men is a reflection of significant admixture with majority populations from the Vardar-Morava-Danube catchment basin - possibly a common paternal signature of Romani populations in Southeastern Europe. Additional traces of admixture are evident in the modest presence of typical European haplogroups.

Are the Moravian Valachs of Czech Republic the Aromuns of Central Europe? Model population for isolation and admixture

E. Ehler1,2, V. Vančata2; 1Department of Anthropology and Human Genetics, Charles University in Prague, Faculty of Science, Prague, Czech Republic, 2Department of Biology and Ecological Education, Charles University in Prague, Faculty of Education, Prague, Czech Republic.

Presentation Number: P07.129

Moravian Valachs of Czech Republic are one of the most distinct ethnic groups from Central Europe. Related to similar populations in Poland and Slovakia, they emerge at the end of 15th century, as the north-westernmost prominence of migration that started 250 years earlier in northern Romania. Being predominately highland sheep herders and of putative Romanian origin, they represent a Central European analogue of Balkan Aromanian populations. We have gathered Y-chromosomal, linguistic, ethnographic and historical data for this population and compared them with surrounding as well as with east European populations. Linguistic data show specific parts of shared vocabulary of Romanian origin between several pastoral groups in Central and Eastern Europe. Comparing genetic and linguistic pairwise distance matrices (Mantel test) in these populations did not revealed any significant correlation. Thus we confirmed that plain geographical distance still plays the major role in genetic distances between populations in Europe. From our further analysis it is clear, that the Moravian Valachs, after at least five centuries of admixture, are not overly genetically different from surrounding populations. On the other hand, from the point of view of intra-population diversity, they are much more similar to isolated Balkan populations (e.g. Aromuns) than to Central European populations.

Phylogeography of the human Y chromosome haplogroup E3a

F. Cruciani1, B. Trombetta1, D. Sellitto2, C. Nodale1, R. Scozzari1; 1Sapienza Università di Roma, Rome, Italy, 2Consiglio Nazionale delle Ricerche, Rome, Italy.

Presentation Number: P07.134

The Y chromosome specific biallelic marker DYS271 defines the most common haplogroup (E3a) currently found in sub-Saharan Africa. A sister clade, E3b (E-M215), is rare in sub-Saharan Africa, but very common in northern and eastern Africa. On the whole, these two clades represent more than 70% of the Y chromosomes of the African continent. A third clade belonging to E3 (E3c or E-M329) has been recently reported to be present only in eastern Africa, at low frequencies. In this study we analyzed more than 1,600 Y chromosomes from 55 African populations, using both new and previously described biallelic markers, in order to refine the phylogeny and the geographic distribution of the E3a haplogroup. The most common E-DYS271 sub-clades (E-DYS271*, E-M191, E-U209) showed a non uniform distribution across sub-Saharan Africa. Most of the E-DYS271 chromosomes found in northern and western Africa belong to the paragroup E-DYS271*, which is rare in central and southern Africa. In these latter regions, haplogroups E-M191 and E-U209 show similar frequency distributions and coalescence ages (13 and 11 kyr, respectively), suggesting their involvement in the same migratory event/s. By the use of two new phylogenetically equivalent markers (V38 and V89), the earlier tripartite structure of E3 haplogroup was resolved in favor of a common ancestor for haplogroups E-DYS271 (formerly E3a) and E-M329 (formerly E3c). The new topology of the E3 haplogroup is suggestive of a relatively recent eastern African origin for the majority of the chromosomes presently found in sub-Saharan Africa.

Y-chromosome lineages in Xhosa and Zulu Bantu speaking populations

R. P. A. Gonçalves, H. Spínola, A. Brehm; Human Genetics Laboratory, Funchal, Portugal.

Presentation Number: P07.137

Y-chromosome Single Nucleotide Polymorphisms have been analysed in Zulu and Xhosa, two southern Africa Bantu speaking populations. These two ethnic groups have their origin on the farmer’s Bantu expansion from Niger-Congo border towards sub-Sahel regions on the southern tip of the continent, during the past 3000 years. Seven different Y-chromosome haplogroups were found in Zulu contrasting with only two in Xhosa. E3a, a common haplogroup among West sub-Saharans associated to Bantu migration was the most prevalent in both populations (56.9% in Zulu and 90% in Xhosa). The second most common haplogroup was E2 (29.3% in Zulu and 10% in Xhosa), present both in West and East African populations. The present-day Zulu and Xhosa paternal legacy is essentially of West sub-Saharan origin. Zulu population shows a most diverse genetic influence comparing to Xhosa, revealing some pre-Bantu expansion markers and East African influences. Zulu presents 8.6% Y-chromosome haplogroups (A, B, J1) of non-Bantu influence that could indicate gene flow from other populations, particularly Khoisan.

Human genetic population structure: Patterns and underlying processes

Presentation Time: Tuesday, 9:15 a.m. - 9:45 a.m.

G. Barbujani; University of Ferrara, Department of Biology and Evolution, Ferrara, Italy.

Presentation Number: S15.2

Classical studies of genetic diversity in humans consistently showed that the largest proportion of human diversity occurs among members of the same population. On average, differences among different populations in the same continent represent 5% of the global human variance, and differences among continents another 10%. Genetic variation is largely discordant across the genome, meaning that different loci show different spatial patterns, and implying that a good description of population structure can only be based on the analysis of multiple loci. Studies of single loci are also unlikely to reasonably identify an individual’s place of origin. A general decline of genetic of genetic diversity with distance from Africa, and a parallel increase in linkage disequilibrium, can be accounted for by the effects of a series of founder effects accompanying the spread of anatomically-modern humans from Africa. Recent DNA analyses at the global level show that most allelic variants are cosmopolitan and only a small percentage are continent-specific, whereas a clearer continental structure emerges when considering composite haplotypes. This suggests that, at the global level, gene flow has had a strong impact on genetic diversity, through both directional dispersal and successive short-range migratory exchanges. At the local level, several factors have contributed to genetic differentiation, and, in particular, language barriers have been shown to be associated with small but non-negligible increases of the genetic differences between neighboring populations.

Hierarchical analysis of 28 Y-chromosome SNP’s in the population of the Republic of Macedonia

P. Noveski, S. Trivodalieva, G. D. Efremov, D. Plaseska-Karanfilska;
Macedonian Academy of Sciences and Arts, Research Centre for Genetic Engineering and Biotechnology, Skopje, Macedonia, The Former Yugoslav Republic of.

Presentation Number: P05.211

Analysis of Y-chromosome haplogroups, defined by single nucleotide polymorphisms (SNP’s), has become a standard approach for studying the origin of human populations and measuring the variability among them. Furthermore, Y-SNP’s represent a new forensic tool, because their population specificity may allow to determine the origin of any male sample of interest for forensic purposes. The aim of this study was to develop a strategy for rapid, simple and inexpensive Y-chromosome SNP’s typing in the population of R. Macedonia. We have studied a total of 343 DNA male samples; 211 Macedonians, 111 Albanians and 21 of other ethnic origin (Roma, Serbs and Turks). Methodology included multiplex PCR and single nucleotide extension reaction by SNaPshot multiplex kit. The set of 28 markers has been grouped in 5 multiplexes in order to determine the most frequent haplogroups using only 1 or 2 multiplexes. Twenty different Y haplogroups were determined among 343 male DNA samples. The finding that five haplogroups (E3b1, I1b1, J2b1a, R1a and R1b) comprise more than 70% of the Y chromosomes is consistent with the typical European Y chromosome gene pool. The distribution of the Y-haplogroups differs between Macedonians and Albanians. The most common Y haplogroup among Macedonians is I1b1 (27.5%), followed by three haplogroups present with similar frequencies E3b1 (15.6%), R1a (14.2%) and R1b (11.4%). Among Albanians the most frequent Y haplogroup is E3b1 (28.8%), followed by R1b (18.0%), J2b1a (13.5%) and R1a (12.6%).

The following paper (probably) refers to a recent study, according to which:

One of the most elevated values of 35delG prevalence corresponds to Greece (1/28); the pattern of various 35delG prevalences is interpretated in the present meta-analysis as the result of Ancient Greek colonizations of the "Magna Grecia" in historical times.

Strong linkage disequilibrium for the frequent GJB2 35delG mutation in the Greek population
H. Kokotas¹, L. Van Laer², M. Grigoriadou¹, V. Iliadou³, J. Economides⁴, S. Pomoni¹, A. Pampanos¹, N. Eleftheriades⁵, E. Ferekidou⁶, S. Korres⁶, A. Giannoulia-Karantana⁷, G. Van Camp², M. B. Petersen¹;
¹Institute of Child Health, Athens, Greece, ²University of Antwerp, Antwerp, Belgium, ³AHEPA Hospital, Thessaloniki, Greece, ⁴‘Aghia Sophia’ Children’s Hospital, Athens, Greece, ⁵St. Loukas Hospital, Thessaloniki, Greece, ⁶Athens University, Athens, Greece, ⁷Athens University Medical School, Athens, Greece.

Presentation Number: P06.080

Approximately one in 1,000 children is affected by severe or profound hearing loss at birth or during early childhood (prelingual deafness). Up to forty percent of autosomal recessive, congenital, severe to profound hearing impairment cases result from mutations in a single gene, GJB2. The 35delG mutation accounts for the majority of GJB2 mutations detected in Caucasian populations and represents one of the most frequent disease mutations identified so far. Some previous studies have assumed that the high frequency of the 35delG mutation reflects the presence of a mutational hot spot, whilst other studies support the theory of a common founder. Greece is amongst the countries presenting high frequency of the 35delG mutation (3.5%), and a recent study raised the hypothesis of the origin of this mutation in ancient Greece. We genotyped 60 Greek deafness patients homozygous for the 35delG mutation for six single nucleotide polymorphisms (SNPs) and two microsatellite markers, mapping within or flanking the GJB2 gene, as compared to 60 Greek hearing controls. A strong linkage disequilibrium was found between the 35delG mutation and markers inside or flanking the GJB2 gene, at distances of 34 kb on the centromeric and 90 kb on the telomeric side of the gene, respectively. Our study supports the hypothesis of a founder effect and we further propose that ethnic groups of Greek ancestry could have propagated the 35delG mutation, as evidenced by historical data beginning from the 15^th century BC.