Andrew J. Pakstis

Global Patterns of Linkage Disequilibrium at the CD4 Locus and Modern Human Origins

Haplotypes consisting of alleles at a short tandem repeat polymorphism (STRP) and an Alu deletion polymorphism at the CD4 locus on chromosome 12 were analyzed in more than 1600 individuals sampled from 42 geographically dispersed populations (13 African, 2 Middle Eastern, 7 European, 9 Asian, 3 Pacific, and 8 Amerindian). Sub-Saharan African populations had more haplotypes and exhibited more variability in frequencies of haplotypes than the Northeast African or non-African populations. The Alu deletion was nearly always associated with a single STRP allele in non-African and Northeast African populations but was associated with a wide range of STRP alleles in the sub-Saharan African populations. This global pattern of haplotype variation and linkage disequilibrium suggests a common and recent African origin for all non-African human populations.

The world-wide distribution of allele frequencies at the human dopamine D4 receptor locus.

Abstract The dopamine D4 receptor gene (DRD4) has an expressed polymorphism in the third exon that may have functional relevance. The polymorphism exists at two levels. At the higher level there is an imperfect tandem repeat of 48 base pairs (bp) coding for 16 amino acids; alleles have been identified with 2 (32 amino acids) to 10 (160 amino acids) repeats. The imperfect nature of the repeats is responsible for a more subtle level of variation since alleles with the same number of repeats can differ in the exact sequences or in the order of the variants of the 48-bp unit. We have undertaken a global survey of this expressed polymorphism as one approach to understanding the evolutionary significance and origins of the polymorphism as well as understanding what selective forces, if any, may be operating at this locus. As the first step, we have determined the repeat number genotype of the DRD4 repeat polymorphism in 1,327 individuals from 36 different populations. The allele frequencies differ considerably among the different populations. The 4-repeat allele was the most prevalent (global mean allele frequency = 64.3%) and appeared in every population with a frequency ranging from 0.16 to 0.96. The 7-repeat allele was the second most common (global mean = 20.6%), appearing quite frequently in the Americas (mean frequency = 48.3%) but only occasionally in East and South Asia (mean frequency = 1.9%). The 2-repeat allele was the third most common (global mean frequency = 8.2%) and was quite frequent in East and South Asia (mean frequency = 18.1%) while uncommon in the Americas (mean frequency = 2.9%) and Africa (mean frequency = 1.7%). The universality of the polymorphism with only three common repeat-number alleles (4, 7, and 2) indicates that the polymorphism is ancient and arose before the global dispersion of modern humans. The diversity of actual allele frequencies for this expressed polymorphism among different populations emphasizes the importance of population considerations in the design and interpretation of any association studies carried out with this polymorphism.

Bitter receptor gene (TAS2R38), 6-n-propylthiouracil (PROP) bitterness and alcohol intake

BACKGROUND Phenylthiocarbamide (PTC) and 6-n-propylthiouracil (PROP), chemically related compounds, are probes for genetic variation in bitter taste, although PROP is safer with less sulfurous odor. Threshold for PROP distinguishes nontasters (increased threshold) from tasters (lower threshold); perceived intensity subdivides tasters into medium tasters (PROP is bitter) and supertasters (PROP is very bitter). Compared with supertasters, nontasters have fewer taste papillae on the anterior tongue (fungiform papillae) and experience less negative (e.g., bitterness) and more positive (eg, sweetness) sensations from alcohol. We determined whether the TAS2R38 gene at 7q36 predicted PROP bitterness, alcohol sensation and use. METHODS Healthy adults (53 women, 31 men; mean age 36 years)--primarily light and moderate drinkers--reported the bitterness of five PROP concentrations (0.032-3.2 mM) and intensity of 50% ethanol on the general Labeled Magnitude Scale. PROP threshold and density of fungiform papillae were also measured. Subjects had common TAS2R38 gene haplotypes [alanine-valine-isoleucine (AVI) and proline-alanine-valine (PAV)]. RESULTS PROP bitterness varied significantly across genotypes with repeated measures ANOVA: 26 AVI/AVI homozygotes tasted less bitterness than either 37 PAV/AVI heterozygotes or 21 PAV/PAV homozygotes. The PAV/PAV group exceeded the PAV/AVI group for bitterness only for the top PROP concentrations. The elevated bitterness was musch less than if we defined the groups using psychophysical criteria. With multiple regression analyses, greater bitterness from 3.2 mM PROP was a significant predictor of greater ethanol intensity and less alcohol intake--effects separate from age and sex. Genotype was a significant predictor of alcohol intake, but not ethanol intensity. With ANOVA, AVI/AVI homozygotes reported higher alcohol use than either PAV/AVI heterozygotes or PAV/PAV homozygotes. When age effects were minimized, PROP bitterness explained more variance in alcohol intake than did the TAS2R38 genotype. CONCLUSIONS These results support taste genetic effects on alcohol intake. PROP bitterness serves as a marker of these effects.

A Global Perspective on Genetic Variation at the ADH Genes Reveals Unusual Patterns of Linkage Disequilibrium and Diversity

Variants of different Class I alcohol dehydrogenase (ADH) genes have been shown to be associated with an effect that is protective against alcoholism. Previous work from our laboratory has shown that the two sites showing the association are in linkage disequilibrium and has identified the ADH1B Arg47His site as causative, with the ADH1C Ile349Val site showing association only because of the disequilibrium. Here, we describe an initial study of the nature of linkage disequilibrium and genetic variation, in population samples from different regions of the world, in a larger segment of the ADH cluster (including the three Class I ADH genes and ADH7). Linkage disequilibrium across approximately 40 kb of the Class I ADH cluster is moderate to strong in all population samples that we studied. We observed nominally significant pairwise linkage disequilibrium, in some populations, between the ADH7 site and some Class I ADH sites, at moderate values and at a molecular distance as great as 100 kb. Our data indicate (1) that most ADH-alcoholism association studies have failed to consider many sites in the ADH cluster that may harbor etiologically significant alleles and (2) that the relevance of the various ADH sites will be population dependent. Some individual sites in the Class I ADH cluster show Fst values that are among the highest seen among several dozen unlinked sites that were studied in the same subset of populations. The high Fst values can be attributed to the discrepant frequencies of specific alleles in eastern Asia relative to those in other regions of the world. These alleles are part of a single haplotype that exists at high (>65%) frequency only in the eastern-Asian samples. It seems unlikely that this haplotype, which is rare or unobserved in other populations, reached such high frequency because of random genetic drift alone.

Linkage disequilibrium at the ADH2 and ADH3 loci and risk of alcoholism.

Two of the three class I alcohol dehydrogenase (ADH) genes (ADH2 and ADH3) encode known functional variants that act on alcohol with different efficiencies. Variants at both these genes have been implicated in alcoholism in some populations because allele frequencies differ between alcoholics and controls. Specifically, controls have higher frequencies of the variants with higher Vmax (ADH2*2 and ADH3*1). In samples both of alcoholics and of controls from three Taiwanese populations (Chinese, Ami, and Atayal) we found significant pairwise disequilibrium for all comparisons of the two functional polymorphisms and a third, presumably neutral, intronic polymorphism in ADH2. The class I ADH genes all lie within 80 kb on chromosome 4; thus, variants are not inherited independently, and haplotypes must be analyzed when evaluating the risk of alcoholism. In the Taiwanese Chinese we found that, only among those chromosomes containing the ADH3*1 variant (high Vmax), the proportions of chromosomes with ADH2*1 (low Vmax) and those with ADH2*2 (high Vmax) are significantly different between alcoholics and controls (P<10-5). The proportions of chromosomes with ADH3*1 and those with ADH3*2 are not significantly different between alcoholics and controls, on a constant ADH2 background (with ADH2*1, P=.83; with ADH2*2, P=.53). Thus, the observed differences in the frequency of the functional polymorphism at ADH3, between alcoholics and controls, can be accounted for by the disequilibrium with ADH2 in this population.

A global survey of haplotype frequencies and linkage disequilibrium at the DRD2 locus

Abstract A four-site haplotype system at the dopamine D2 receptor locus (DRD2) has been studied in a global sample of 28 distinct populations. The haplotype system spans about 25 kb, encompassing the coding region of the gene. The four individual markers include three TaqI restriction site polymorphisms (RSPs) – TaqI “A”, “B”, and “D” sites – and one dinucleotide short tandem repeat polymorphism (STRP). All four of the marker systems are polymorphic in all regions of the world and in most individual populations. The haplotype system shows the highest average heterozygosity in Africa, a slightly lower average heterozygosity in Europe, and the lowest average heterozygosities in East Asia and the Americas. Across all populations, 20 of the 48 possible haplotypes reached a frequency of at least 5% in at least one population sample. However, no single population had more than six haplotypes reaching that frequency. In general, African populations had more haplotypes present in each population and more haplotypes occurring at a frequency of at least 5% in that population. Permutation tests for significance of overall disequilibrium (all sites considered simultaneously) were highly significant (P<0.001) in all 28 populations. Except for three African samples, the pairwise disequilibrium between the outermost RSP markers, TaqI “B” and “A”, was highly significant with D’ values greater than 0.8; in two of those exceptions the RSP marker was not polymorphic. Except for those same two African populations, the 16-repeat allele at the STRP also showed highly significant disequilibrium with the TaqI “B” site in all populations, with D’ values usually greater than 0.7. Only four haplotypes account for more than 70% of all chromosomes in virtually all non-African populations, and two of those haplotypes account for more than 70% of all chromosomes in most East Asian and Amerindian populations. A new measure of the amount of overall disequilibrium shows least disequilibrium in African populations, somewhat more in European populations, and the greatest amount in East Asian and Amerindian populations. This pattern seems best explained by random genetic drift with low levels of recombination, a low mutation rate at the STRP, and essentially no recurrent mutation at the RSP sites, all in conjunction with an “Out of Africa” model for recent human evolution.

Obsessive-compulsive symptom dimensions in affected sibling pairs diagnosed with Gilles de la Tourette syndrome.

Obsessive‐compulsive disorder (OCD) is an etiologically heterogeneous disorder. Recent factor analyses have consistently identified several symptom dimensions, two of which are associated with increased familial risk for OCD; aggressive, sexual, and religious obsessions and checking compulsions (FACTOR 1) and symmetry and ordering obsessions and compulsions (FACTOR 2). Both of these symptom dimensions are also frequently seen in association with Gilles de la Tourette syndrome (GTS). The purpose of this study was to determine whether these obsessive‐compulsive (OC) symptom dimensions are correlated within families (between sibs and between parent‐child pairs). Using data collected by the Tourette Syndrome Association International Consortium for Genetics Affected Sibling Pair Study, the authors selected all available GTS sib pairs and their parents for which these OC symptom dimensions (factor scores) could be generated. This group included 128 full sibs and their mothers (54) and fathers (54). Four OC symptom dimension scores were computed for each family member using an algorithm derived from item endorsements from the Yale‐Brown Obsessive‐Compulsive Scale (Y‐BOCS) symptom checklist. In addition to a series of univariate analyses, complex segregation analyses were also completed using these quantitative OC symptom dimension scores. FACTOR 1 and FACTOR 2 scores were significantly correlated in sib pairs concordant for GTS. The mother–child correlations, but not father–child correlations, were also significant for these two factors. Segregation analyses were consistent with dominant major gene effects for both FACTOR 1 and FACTOR 2. We conclude that familial factors contribute significantly to OC symptom dimension phenotypes in GTS families. This familial contribution could be genetic or environmental.

Haplotypes and linkage disequilibrium at the phenylalanine hydroxylase locus, PAH, in a global representation of populations.

Because defects in the phenylalanine hydroxylase gene (PAH) cause phenylketonuria (PKU), PAH was studied for normal polymorphisms and linkage disequilibrium soon after the gene was cloned. Studies in the 1980s concentrated on European populations in which PKU was common and showed that haplotype-frequency variation exists between some regions of the world. In European populations, linkage disequilibrium generally was found not to exist between RFLPs at opposite ends of the gene but was found to exist among the RFLPs clustered at each end. We have now undertaken the first global survey of normal variation and disequilibrium across the PAH gene. Four well-mapped single-nucleotide polymorphisms (SNPs) spanning approximately 75 kb, two near each end of the gene, were selected to allow linkage disequilibrium across most of the gene to be examined. These SNPs were studied as PCR-RFLP markers in samples of, on average, 50 individuals for each of 29 populations, including, for the first time, multiple populations from Africa and from the Americas. All four sites are polymorphic in all 29 populations. Although all but 5 of the 16 possible haplotypes reach frequencies >5% somewhere in the world, no haplotype was seen in all populations. Overall linkage disequilibrium is highly significant in all populations, but disequilibrium between the opposite ends is significant only in Native American populations and in one African population. This study demonstrates that the physical extent of linkage disequilibrium can differ substantially among populations from different regions of the world, because of both ancient genetic drift in the ancestor common to a large regional group of modern populations and recent genetic drift affecting individual populations.

The evolution and population genetics of the ALDH2 locus: random genetic drift, selection, and low levels of recombination

The catalytic deficiency of human aldehyde dehydrogenase 2 (ALDH2) is caused by a nucleotide substitution (G1510A; Glu487Lys) in exon 12 of the ALDH2 locus. This SNP, and four non‐coding SNPs, including one in the promoter, span 40 kb of ALDH2; these and one downstream STRP have been tested in 37 worldwide populations. Only four major SNP‐defined haplotypes account for almost all chromosomes in all populations. A fifth haplotype harbours the functional variant and is only found in East Asians. Though the SNPs showed virtually no historic recombination, LD values are quite variable because of varying haplotype frequencies, demonstrating that LD is a statistical abstraction and not a fundamental aspect of the genome, and is not a function solely of recombination. Among populations, different sets of tagging SNPs, sometimes not overlapping, can be required to identify the common haplotypes. Thus, solely because haplotype frequencies vary, there is no common minimum set of tagging SNPs globally applicable. The Fst values of the promoter region SNP and the functional SNP were about two S.D. above the mean for a reference distribution of 117 autosomal biallelic markers. These high Fst values may indicate selection has operated at these or very tightly linked sites.

Analyses of a set of 128 ancestry informative single-nucleotide polymorphisms in a global set of 119 population samples

BackgroundUsing DNA to determine an individuals ancestry from among human populations is generally interesting and useful for many purposes, including admixture mapping, controlling for population structure in disease or trait association studies and forensic ancestry inference. However, to estimate ancestry, including possible admixture within an individual, as well as heterogeneity within a group of individuals, allele frequencies are necessary for what are believed to be the contributing populations. For this purpose, panels of ancestry informative markers (AIMs) have been developed.ResultsWe are presenting our work on one such panel, composed of 128 ancestry informative single-nucleotide polymorphisms (AISNPs) already proposed in the literature. Compared to previous studies of these AISNPs, we have studied three times the number of individuals (4,871) in three times as many population samples (119). We have validated this panel for many ancestry assignment and admixture studies, especially those that were the rationale for the original selection of the 128 SNPs: African Americans and Mexican Americans. At the same time, the limitations of the panel for distinguishing ancestry and quantifying admixture among Eurasian populations are noted.ConclusionWe demonstrate the simultaneous importance of the specific set of population samples and their relative sample sizes in the use of the structure program to determine which groups cluster together and consequently influence the ability of a marker panel to infer ancestry. We demonstrate the strengths and weaknesses of this particular panel of AISNPs in a global context.