Native Americans have been divided into three linguistic groups: the reasonably well-defined Eskaleut and Nadene of northern North America and the highly heterogeneous Amerind of North, Central, and South America. The heterogeneity of the Amerinds has been proposed to be the result of either multiple independent migrations or a single ancient migration with extensive in situ radiation, To investigate the origin and interrelationship of the American Indians, we examined the mitochondrial DNA (mtDNA) variation in 87 Amerinds (Pima, Maya, and Ticuna of North, Central, and South America, respectively), 80 Nadene (Dogrib and Tlingit of northwest North America and Navajo of the southwest North America), and 153 Asians from 7 diverse populations. American Indian mtDNAs were found to be directly descended from five founding Asian mtDNAs and to cluster into four lineages, each characterized by a different rare Asian mtDNA marker. Lineage A is defined by a HaeIII site gain at np 663, lineage B by a 9-bp deletion between the COII and tRNA^Lys genes, lineage C by a HincII site loss at np 13259, and lineage D by an AluI site loss at np 5176. The North, Central, and South America Amerinds were found to harbor all four lineages, demonstrating that the Amerinds originated from a common ancestral genetic stock. The genetic variation of three of the four Amerind lineages (A, C, and D) was similar with a mean value of 0.084%, whereas the sequence variation in the fourth lineage (B) was much lower, raising the possibility of an independent arrival. By contrast, the Nadene mtDNAs were predominantly from lineage A, with 27% of them having a Nadene-specific RsaI site loss at np 16329. The accumulated Nadene variation was only 0.021%. These results demonstrate that the Amerind mtDNAs arose from one or maybe two Asian migrations that were distinct from the migration of the Nadene and that the Amerind populations are about four times older than the Nadene.

The genetic determinants of physiological and developmental responses to environmental stress are poorly understood, This has been primarily due to the difficulty of direct measurement of response and the lack of appropriate statistical genetic methods. Here, I present a unified statistical genetic methodology for human adaptability studies that permits evaluation of the inheritance of quantitative trait response to environmental stressors. The foundation of this approach is the mathematical relationship between genotype-environment interaction and the genetic variance of response to environmental challenge. I describe two basic methods that can be used for either discrete or continuous environments. Each method allows for major loci, residual polygenic variation, and genotype-environment interaction at both the major genic and the polygenic levels. The first method is based on multivariate segregation analysis and is appropriate for situations in which data are available for each individual in each environment. The second method is appropriate for the more common case when response to the environment cannot be observed directly. This method is based on an extension of a mixed major locus/variance component model and can be used when singly measured related individuals are observed in different environments. Three example applications using data on lipoprotein variation in pedigreed baboons are provided to show the utility of these methods.

We examine the associations between nutrition and mortality at the national level. Altogether four aspects of this association are explored: (1) total calories with expectation of life, (2) dietary composition with expectation of life, (3) total calories with the age patterns of mortality, and (4) dietary composition with the age patterns of mortality. The data consist of life tables and national food balance sheets for 341 populations from 96 countries. A preliminary principal components analysis conducted on the dietary composition data yields three dietary components: (1) the overall quality and quantity of the diet, (2) the relative contribution of carbohydrates versus fats, and (3) the relative contribution of fats versus proteins. The results indicate that expectation of life at birth increases with total calories, with overall quality and quantity of the diet, and with the ratio of fats to proteins. The ratio of carbohydrates to fats is negatively associated with level of mortality. However, evidence indicates that the main effect of the ratio of fats to proteins is reversed when diets are high in quality and that all the effects tend to saturate at high nutrient availability. Variation in nutrition is also strongly associated with the international variation in age patterns of mortality. For example, when expectation of life is held constant, populations with higher quality diets tend to have lower childhood mortality and higher adult mortality. The results indicate that nutritional patterns are highly correlated with much of the worldwide variation in mortality and may be a useful criterion for selecting or predicting the best suited model life table for use on a particular population.

The use of multivariate quantitative trait information to address questions of population relationships and evolutionary issues has a longstanding history in human anthropometry. Previous analyses have usually rested on a number of explicit or implicit assumptions that allow phenotypic information to be used as a proxy for quantitative genetic information. One (usually implicit) assumption is that the additive genetic variance-covariance matrix (G) among traits is proportional to the phenotypic variance-covariance matrix (P). In this study we discuss the implications of this assumption, demonstrating that if it is true that G = h²P, where h² is some constant of proportionality, then (1) the biological (phenotypic) Mahalanobis distance will be proportional to genetic distance, (2) phenotypic and genetic allometry coefficients will be equal, and (3) evolutionary models will become simplified. We then use a multivariate quantitative genetic analysis of 12 anthropometric traits in 5 tribes to demonstrate that G = h²P for at least a portion of the Boas data.

In most of the world's population the ability to digest lactose declines sharply after infancy. High lactose digestion capacity in adults is common only in populations of European and circum-Mediterranean origin and is thought to be an evolutionary adaptation to millennia of drinking milk from domestic livestock. Milk can also be consumed in a processed form, such as cheese or soured milk, which has a reduced lactose content. Two other selective pressures for drinking fresh milk with a high lactose content have been proposed: promotion of calcium uptake in high-latitude populations prone to vitamin-D deficiency and maintenance of water and electrolytes in the body in highly arid environments. These three hypotheses are all supported by the geographic distribution of high lactose digestion capacity in adults. However, the relationships between environmental variables and adult lactose digestion capacity are highly confounded by the shared ancestry of many populations whose lactose digestion capacity has been tested. The three hypotheses for the evolution of high adult lactose digestion capacity are tested here using a comparative method of analysis that takes the problem of phylogenetic confounding into account. This analysis supports the hypothesis that high adult lactose digestion capacity is an adaptation to dairying but does not support the hypotheses that lactose digestion capacity is additionally selected for either at high latitudes or in highly arid environments. Furthermore, methods using maximum likelihood are used to show that the evolution of milking preceded the evolution of high lactose digestion.

Allele frequencies are clinally distributed for many protein polymorphisms in Europe, suggesting that the current populations are derived from an ancestral group that expanded from the Near East. It is not yet fully established whether that expansion took place during the Neolithic or earlier or whether the detectable protein variation faithfully reflects the underlying molecular variation. In this study we address the latter question by describing geographic patterns of genetic diversity at seven highly polymorphic DNA markers. Two of these markers are minisatellites, four are microsatellites, and the seventh is a locus of the HLA system. By analyzing a database of 304 samples, with more than 130,000 chromosomes, we found evidence for a major clinal component of genetic variation. At most loci spatially close populations resemble each other genetically, and the degree of genetic similarity, as measured by spatial autocorrelation statistics, decreases at increasing distances. The observed patterns of molecular variation do not seem to differ qualitatively from those identified for protein polymorphisms. This suggests that low levels of population structuring, described in some mitochondrial DNA studies, may reflect different evolutionary histories for nuclear and maternally inherited markers or, alternatively, that spatial patterns of mitochondrial DNA variation may need more sensitive statistical methods to be recognized.

High frequencies of some rare inherited recessive disorders can be found in the Saguenay region of Quebec, Canada. Four disorders have a carrier frequency of about 0.04 (in the range 0.035–0.05): pseudo—vitamin D—dependent rickets, hereditary tyrosinemia type 1, Charlevoix-Saguenay spastic ataxia, and sensorimotor polyneuropathy with or without agenesis of the corpus callosum. Molecular data suggest that only 1 mutation has been introduced into the population since its founding in the 17th century. The carrier frequencies are much higher than one would expect under a theoretical model that includes variance in family size and population growth (Thompson and Neel 1978). I present a methodology called allele dropping to test the hypothesis that only 1 founder introduced a given mutation. This study is based on 891 ascending genealogies and enables one to measure the extent of allele frequency changes resulting from the demographic history of the population. Two scenarios are tested: neutral and lethal alleles. Lethality has a minor effect because the alleles never reach a frequency high enough for selection to be strong. Twenty-five founders have a probability greater than 1% that a lethal mutation they introduced into the population will reach a carrier frequency between 0.035 and 0.05 in the contemporary population. Moreover, 2 founders have a probability greater than 20% that a lethal allele they introduced into the population will reach this target frequency. Therefore the simplest hypothesis that 1 founder introduced 1 disorder into the population is consistent.

In the 20th century geneticists began to unravel some of the simpler aspects of the etiology of inherited diseases in humans. The theory of linkage analysis was developed and applied long before the advent of molecular biology, but only the technological advances of the second half of the 20th century made large-scale gene mapping with a dense genome-spanning set of markers a reality. More recently, the primary topic of interest has shifted from simple Mendelian diseases, for which genotypes of some gene are the cause of disease, to more complex diseases, for which genotypes of some set of genes together with environmental factors merely alter the probability that an individual gets the disease, although individual factors are typically insufficient to cause the disease outright. To this end, a great deal of dogma has evolved about the best way to skin this cat, although to date success has been minimal with any approach. We postulate that the main reason for this is a lack of attention to experimental design. Once the data have been ascertained, the most powerful statistical methods will not be able to salvage an inappropriately designed study (Andersen 1990). Each phenotype and/or population mandates its own individually tailored study design to maximize the chances of successful gene mapping. We suggest that careful consideration of the available data from real genotype-phenotype correlation studies (as opposed to oversimplified theoretically tractable models), and the practical feasibility of different ascertainment schemes dictate how one should proceed. In this review we review the theory and practice of gene mapping at the close of the 20th century, showing that most methods of linkage and linkage disequilibrium analysis are similar in a fundamental sense, with the differences being related more to study design and ascertainment than to technical details of the underlying statistical analysis. To this end, we propose a new focus in the field of statistical genetics that more explicitly highlights the primacy of study design as the means to increase power for gene mapping.

In recent years multilocus data sets have been used to study the demographic history of human populations. In this paper (1) analyses previously done on 60 short tandem repeat (STR) loci are repeated on 30 restriction site polymorphism (RSP) markers; (2) relative population weights are estimated from the RSP data set and compared to previously published estimates from STR and craniometric data sets; and (3) computer simulations are performed to show the effects of ascertainment bias on relative population weight estimates. Not surprisingly, given that the RSP markers were originally identified in a small panel of Caucasians, estimates of relative population weights are biased and the European population weight is artificially inflated. However, the effects of ascertainment bias are not apparent in a principal components plot or estimates of F_ST. Ascertainment bias can have a large effect in other genetic systems with inherently low heterozygosity such as Alus or single nucleotide polymorphisms (SNPs), and care must be taken to have prior knowledge of how polymorphic markers in a given data set were originally identified. Otherwise, results can be skewed and interpretations faulty.

Birth-weight-specific infant mortality is examined using a novel statistical procedure, parametric mixtures of logistic regressions. The results indicate that birth cohorts are composed of two or more subpopulations that are heterogeneous with respect to infant mortality. One subpopulation appears to account for the “normal” process of fetal development, while the other, which accounts for the majority of births at both low and high birth weights, may represent fetuses that were “disturbed” during development. Surprisingly, estimates of neonatal and infant mortality indicate that the “disturbed” subpopulation has lower birth-weight-specific mortality, although overall crude mortality rates are higher for this subpopulation. It is hypothesized that this is due to high rates of fetal loss among the “disturbed” subpopulation, resulting in a highly selected group at birth. The heterogeneity identified in the birth cohort could be responsible for recent decelerations in the decline in infant mortality, and might be the cause of unexplained ethnic differences in birth-weight-specific infant mortality. The novel statistical methodology developed here has broad application within human biology. In particular, it could be used in any context where parametric mixture modeling is applied, such as complex segregation analysis.

Sewall Wright's population structure statistic, F_ST, measured among samples of world populations is often 15% or less. This would indicate that 85% of genetic variation occurs within groups while only 15% can be attributed to allele frequency differences among groups. In this paper, we show that this low value reflects strong biases that result from violating hidden assumptions that define F_ST. These limitations on F_ST are demonstrated algebraically and in the context of analyzing dinucleotide repeat allele frequencies for a set of eight loci genotyped in eight human groups and in chimpanzees. In our analyses, estimates of F_ST. fail to identify important variation. For example, when the analysis includes only humans, F_ST = 0.119, but adding the chimpanzees increases it only a little, F_ST = 0.183. By relaxing the underlying statistical assumptions, the results for chimpanzees become consistent with common knowledge, and we see a richer pattern of human genetic diversity. Some human groups are far more diverged than would be implied by standard computations of F_ST, while other groups are much less diverged. We discuss the relevance of these findings to the application of biological race concepts to humans. Four different race concepts are considered: typological, population, taxonomic, and lineage. Surprisingly, a great deal of genetic variation within groups is consistent with each of these concepts. However, none of the race concepts is compatible with the patterns of variation revealed by our analyses.

A primary objection from a population genetics perspective to a multiregional model of modern human origins is that the model posits a large census size, whereas genetic data suggest a small effective population size. The relationship between census size and effective size is complex, but arguments based on an island model of migration show that if the effective population size reflects the number of breeding individuals and the effects of population subdivision, then an effective population size of 10,000 is inconsistent with the census size of 500,000 to 1,000,000 that has been suggested by archeological evidence. However, these models have ignored the effects of population extinction and recolonization, which increase the expected variance among demes and reduce the inbreeding effective population size. Using models developed for population extinction and recolonization, we show that a large census size consistent with the multiregional model can be reconciled with an effective population size of 10,000, but genetic variation among demes must be high, reflecting low interdeme migration rates and a colonization process that involves a small number of colonists or kin-structured colonization. Ethnographic and archeological evidence is insufficient to determine whether such demographic conditions existed among Pleistocene human populations, and further work needs to be done. More realistic models that incorporate isolation by distance and heterogeneity in extinction rates and effective deme sizes also need to be developed. However, if true, a process of population extinction and recolonization has interesting implications for human demographic history.

Although much attention has been given to statistical genetic methods for the initial localization and fine mapping of quantitative trait loci (QTLs), little methodological work has been done to date on the problem of statistically identifying the most likely functional polymorphisms using sequence data. In this paper we provide a general statistical genetic framework, called Bayesian quantitative trait nucleotide (BQTN) analysis, for assessing the likely functional status of genetic variants. The approach requires the initial enumeration of all genetic variants in a set of resequenced individuals. These polymorphisms are then typed in a large number of individuals (potentially in families), and marker variation is related to quantitative phenotypic variation using Bayesian model selection and averaging. For each sequence variant a posterior probability of effect is obtained and can be used to prioritize additional molecular functional experiments. An example of this quantitative nucleotide analysis is provided using the GAW12 simulated data. The results show that the BQTN method may be useful for choosing the most likely functional variants within a gene (or set of genes). We also include instructions on how to use our computer program, SOLAR, for association analysis and BQTN analysis.

Localization of human quantitative trait loci (QTLs) is now routine. However, identifying their functional DNA variants is still a formidable challenge. We present a complete dissection of a human QTL using novel statistical techniques to infer the most likely functional polymorphisms of a QTL that influence plasma levels of clotting factor VII (FVII), a risk factor for cardiovascular disease. Resequencing of 15 kb in and around the F7 gene identified 49 polymorphisms, which were then genotyped in 398 people. Using a Bayesian quantitative trait nucleotide (BQTN) method, we identified four to seven functional variants that completely account for this QTL. These variants include both rare coding variants and more common, potentially regulatory polymorphisms in intronic and promoter regions.

Most of the inhabitants of South Tyrol in the eastern Italian Alps can be considered isolated populations because of their physical separation by mountain barriers and their socioculturel heritage. We analyzed the genetic structure of South Tyrolean populations using three types of genetic markers: Y-chromosome, mitochondrial DNA (mtDNA), and autosomal Alu markers. Using random samples taken from the populations of Val Venosta, Val Pusteria, Val Isarco, Val Badia, and Val Gardena, we calculated genetic diversity within and among the populations. Microsatellite diversity and unique event polymorphism diversity (on the Y chromosome) were substantially lower in the Ladin-speaking population of Val Badia compared to the neighboring German-speaking populations. In contrast, the genetic diversity of mtDNA haplotypes was lowest for the upper Val Venosta and Val Pusteria. These data suggest a low effective population size, or little admixture, for the gene pool of the Ladin-speaking population from Val Badia. Interestingly, this is more pronounced for Ladin males than for Ladin females. For the pattern of genetic Alu variation, both Ladin samples (Val Gardena and Val Badia) are among the samples with the lowest diversity. An admixture analysis of one German-speaking valley (Val Venosta) indicates a relatively high genetic contribution of Ladin origin. The reduced genetic diversity and a high genetic differentiation in the Rhaetoroman-and German-speaking South Tyrolean populations may constitute an important basis for future medical genetic research and gene mapping studies in South Tyrol.

Human mannose-binding lectin (MBL) plays an important role in innate immunity. MBL deficiency is associated with mutations in the promoter region and in exon 1 of the MBL2 gene. Such deficiency has been correlated with elevated incidence of infections in infancy and in immunocompromised adults. We determined the distribution profile of the MBL2 gene variants in the general population of Benin (West Africa) and in a vulnerable subset of children with sickle cell disease (SCD) (SS homozygotes). Five hundred forty-two healthy individuals (274 newborns, 268 adults) and 128 patients with SCD (35 newborns, 93 children) were screened for the common variant alleles in the MBL2 secretor haplotype region (exon 1 and promoter). The p.G57E variant allele was the most frequent allele compared to p.G54D (27.5% vs. 1.6%, respectively). The p.R52C allele was not found in this population. There was no difference in allele or genotype frequencies between healthy newborns and newborns with SCD. Alleles associated with MBL deficiency were more frequent in adults than in newborns (69.8% vs. 57.3%, respectively; p = 0.002). This enrichment was exclusively due to an elevated proportion of heterozygotes for the p.G57E allele (47.0% vs. 35.3%, respectively; p = 0.004), supporting a potential selective advantage of this genotype. Our results, compared to those reported in other African countries, support the implication of the MBL2 gene in various major infections in Africa, such as meningitis and tuberculosis in HIV-positive patients.

The ρ statistic is commonly used to infer chronological dates for molecular lineages, especially from mitochondrial DNA sequences obtained in anthropological contexts. Since this approach was described 12 years ago, it has been applied to estimate molecular dates in more than 200 studies, including some published in top-tier journals. However, this method has not been well evaluated, and the accuracy of dates obtained from the ρ statistic remains unknown, especially for genetic data collected from populations with complex demographic histories. Here, molecular dates inferred from ρ are compared against coalescent expectations from a range of demographic models. This exercise reveals considerable inaccuracy. Molecular dates based on ρ have a slight downward bias with large asymmetric variance and commonly exhibit substantial type I error rates, where the true age of a lineage falls outside the 95% confidence bounds derived from the variance of ρ. Furthermore, demography proves to be a strong confounding factor in estimating molecular dates accurately, especially for populations in which bottlenecks, founder events, and size changes have played important historical roles. Therefore considerable caution should be applied to inferences made from molecular dates based on the ρ statistic, many of which may be misleading and warrant considerable skepticism.