A novel measure of genetic distance for highly polymorphic tandem repeat loci

Genetic distance measures are indicators of relatedness among populations or species and are useful for reconstructing the historic and phylogenetic relationships among such groups. Classical measures of genetic distance were developed to analyze biochemical and serological polymorphisms, systems which generally show limited variability. However, these traditional measures of genetic distance are inadequate for the analysis of certain classes of variable number tandem repeat (VNTR) loci, which have a larger number of alleles and higher levels of heterozygosity than traditional genetic markers. At the higher levels of heterozygosity observed at these loci, the standard measures of genetic distance are nonlinear and do not account for the mutational mechanisms of hypervariable loci. We have developed a measure of genetic distance, DSW, which is appropriate for the analysis of highly polymorphic DNA loci. Using computer simulations of diverging populations, we show that DSW conforms to linearity and that the variance is similar in magnitude to traditional measures of genetic distance. Comparisons of phylogenetic trees derived from the simulated divergence of human racial groups demonstrate that the branch lengths of trees prepared using DSW are more similar to the model tree than those generated using other measures. Finally, we demonstrate the applicability of DSW to evolutionary analysis by reconstructing the relationships among eight human populations using 14 microsatellite and STR loci. The phylogenetic trees generated using DSW are different from trees constructed with traditional measures and better reflect the well-documented ancient divergence of African and non-African populations.

Intra- and inter-population diversity at short tandem repeat loci in diverse populations of the world

To study the level of intra- and inter-population variation at hypervariable DNA loci, we have characterized 15 human populations of diverse ethnic and geographic origins at six short tandem repeat loci by using the polymerase chain reaction. Even though the spectrum of allelic variation is quite broad and there are substantial differences in allele frequency distributions among populations, in general, population within a major racial group show a greater degree of similarity. This observation is reflected in the analysis of gene diversity. When the total diversity is apportioned, the maximum variation becomes attributable to inter-individual differences within a population; of the variation that is attributed to differences between populations within a racial group and differences between racial groups, namely, African, Caucasian, and Mongoloid, than the American Indians and the Pacific Islanders. As expected, a reciprocal relationship between gene diversity and FST levels is observed. Higher values of FST in the American Indian and the Pacific Islanders may reflect smaller population size and a higher level of isolation. An analysis of genetic distance encompassing the populations belonging to the three major racial groups recognizes three distinct clusters - all the populations of African affiliation cluster together, as do the Caucasian affiliated and the Mongoloid groups, in two distinct clusters. Interestingly, three broadly classified cosmopolitan US populations, namely, US White, US Black and US Asian, cluster with their ancestrally related populations. This study dispels some of the concerns regarding the applicability of DNA typing data for forensic use.

Cancer predisposition, radiosensitivity and the risk of radiation-induced cancers. I. Background

This paper presents an overview of current knowledge on genetic predisposition to cancer and on enhanced sensitivity of cancer-predisposed genotypes to cancers induced by ionizing radiation. It is intended to provide a background and set the stage for the next papers in this series in which we will assess how such heterogeneity (with respect to predisposition to cancer and presence of radiosensitivity genotypes) in a population may affect estimates of the risk of radiation-induced cancers. The main findings and/or conclusions of the present paper are the following: (1) "Cancer-predisposing genes" (i.e. those at which germinal mutations predispose to cancer) are present in the human genome; these genes are responsible not only for the rare familial cancer syndromes but also for a proportion of the common cancers. At least 21 such genes have now been cloned (including 9 tumor suppressor genes, 11 DNA repair genes and 1 proto-oncogene); further, at least 8 putative tumor suppressor genes and a gene involved in ataxia telangiectasia have been localized to specific chromosomes. (2) These genes play crucial roles in the control of cellular proliferation, programmed cell death (apoptosis) and/or one or another DNA repair pathway. Consequently, mutations in these genes are likely to "liberate" the cells from the normal constraints imposed by them, resulting in unconstrained growth characteristic of cancer. (3) At present, the evidence for increased sensitivity of cancer-predisposed genotypes to radiation-induced cancers is limited. However, current knowledge of the known functions of the cancer-predisposing genes and of the consequences of mutations in these provide (a) sufficient grounds for assuming that the genotypes of those predisposed to cancer may be at an increased risk for radiation-induced cancers and (b) the rationale for attempts to estimate quantitatively the impact of genotype-dependent differences in cancer predisposition and radiosensitivity on cancer risks in an irradiated population.

Phylogeny of six Sciurus aberti subspecies based on nucleotide sequences of cytochrome b

The tassel-eared squirrel, Sciurus aberti, is separated into six subspecies which occupy restricted and comparable habitats in ponderosa pine forests in the south-western United States and Mexico. These forests and squirrel populations are currently isolated by large arid areas and, as such, S. aberti appears to offer an example of incipient speciation. We sequenced the complete mitochondrial cytochrome b gene to construct a molecular phylogeny for S. aberti and to determine whether subspecific genetic structure and geographic patterns are correlative. Twenty alleles were identified among 612 squirrels throughout the species' range. Nucleotide divergence between alleles ranged from 0.009 to 0.0233, whereas average sequence divergence between S. aberti and an outgroup species, Sciurus niger, was 0.1823. Neighbor-joining and maximum parsimony analyses identified three major genetic assemblages composed of the following subspecies groups: (1) barberi and durangi; (2) aberti and kaibabensis; and (3) ferreus. The chuscensis samples were the only population with divergent sequences; one sequence was identical to an aberti sequence and a second unique sequence clustered with the ferreus sequences. The presence of divergent sequences in the chuscensis population, coupled with its central geographic position between aberti and ferreus, suggests a relatively recent influx of aberti mtDNA. Estimates of the times separating sequences in subspecies within different groups ranged from 0.94 to 1.52 x 10(6) years, based on a rate estimate of 7.15 x 10(-9) substitutions/year/site. The limited divergence observed between (1) aberti and kaibabensis as well as (2) barberi and durangi suggests relatively recent separations of subspecies within each assemblage. In fact, populations defined morphologically and geographically as ferreus exhibited greater sequence divergence than the aforementioned groups, identifying ferreus as the subspecies with the greatest genetic substructuring. The levels of cytochrome b divergence observed for the three distinct groups argues against a significant role for late Pleistocene glaciation in dispersal of this particular species; however, the proposed intermixing of aberti and chuscensis populations may well have been associated with such glacial events. Nucleotide diversity within subspecies ranked chuscensis >> aberti > barberi approximately kaibabensis approximately ferreus subpopulations; the relatively high level of diversity of chuscensis samples likely results from the apparent introgression of an aberti haplotype. The comparative levels of diversity in the aberti, barberi, kaibabensis, and ferreus sample populations do not correlate with respective habitat size (and presumably population size), suggesting that relatively recent forces, e.g., glaciation and inconsistent timber harvests, may have influenced diversity in these populations without apparent alterations in population size.

Phylogeny of six Sciurus aberti subspecies based on nucleotide sequences of cytochrome b

The tassel-eared squirrel, Sciurus aberti, is separated into six subspecies which occupy restricted and comparable habitats in ponderosa pine forests in the south-western United States and Mexico. These forests and squirrel populations are currently isolated by large arid areas and, as such, S. aberti appears to offer an example of incipient speciation. We sequenced the complete mitochondrial cytochrome b gene to construct a molecular phylogeny for S. aberti and to determine whether subspecific genetic structure and geographic patterns are correlative. Twenty alleles were identified among 612 squirrels throughout the species' range. Nucleotide divergence between alleles ranged from 0.009 to 0.0233, whereas average sequence divergence between S. aberti and an outgroup species, Sciurus niger, was 0.1823. Neighbor-joining and maximum parsimony analyses identified three major genetic assemblages composed of the following subspecies groups: (1) barberi and durangi; (2) aberti and kaibabensis; and (3) ferreus. The chuscensis samples were the only population with divergent sequences; one sequence was identical to an aberti sequence and a second unique sequence clustered with the ferreus sequences. The presence of divergent sequences in the chuscensis population, coupled with its central geographic position between aberti and ferreus, suggests a relatively recent influx of aberti mtDNA. Estimates of the times separating sequences in subspecies within different groups ranged from 0.94 to 1.52 x 10(6) years, based on a rate estimate of 7.15 x 10(-9) substitutions/year/site. The limited divergence observed between (1) aberti and kaibabensis as well as (2) barberi and durangi suggests relatively recent separations of subspecies within each assemblage. In fact, populations defined morphologically and geographically as ferreus exhibited greater sequence divergence than the aforementioned groups, identifying ferreus as the subspecies with the greatest genetic substructuring. The levels of cytochrome b divergence observed for the three distinct groups argues against a significant role for late Pleistocene glaciation in dispersal of this particular species; however, the proposed intermixing of aberti and chuscensis populations may well have been associated with such glacial events. Nucleotide diversity within subspecies ranked chuscensis >> aberti > barberi approximately kaibabensis approximately ferreus subpopulations; the relatively high level of diversity of chuscensis samples likely results from the apparent introgression of an aberti haplotype. The comparative levels of diversity in the aberti, barberi, kaibabensis, and ferreus sample populations do not correlate with respective habitat size (and presumably population size), suggesting that relatively recent forces, e.g., glaciation and inconsistent timber harvests, may have influenced diversity in these populations without apparent alterations in population size.

Correlation of DNA fragment sizes within loci in the presence of non-detectable alleles

At present most forensic databases of DNA profiling of individuals consist of DNA fragment sizes measured from Southern blot restriction fragment length polymorphism (RFLP) analysis. Statistical studies of these databases have revealed that, when fragment sizes are measured from RFLP analysis, some of the single-band patterns of individuals may actually be due to heterozygosity of alleles in which fragment size resulting from one allele remains undetected. In this work, we evaluate the effect of such allelic non-detectability on correlation of fragment sizes within individuals at a locus, and its impact on the inference of independence of fragment sizes within loci. We show that when non-detectable alleles are present in a population at a locus, positive correlations of fragment sizes are expected, which increase with the proportion of non-detectable alleles at the locus. Therefore, a non-zero positive correlation is not a proof of allelic dependence within individuals. Applications of this theory to the current forensic RFLP databases within the US show that there is virtually no evidence of significant allelic dependence within any of the loci. Therefore, the assumption that DNA fragment sizes within loci are independent is valid, and hence, the population genetic principles of computing DNA profile frequencies by multiplying binned frequencies of fragment sizes are most likely to be appropriate for forensic applications of DNA typing data.

Population structure, stepwise mutations, heterozygote deficiency and their implications in DNA forensics

In a substructured population the overall heterozygote deficiency can be predicted from the number of subpopulations (s), their time of divergence (t), and the nature of the mutations. At present the true mutational mechanisms at the hypervariable DNA loci are not known. However, the two existing mutation models (the infinite allele model (IAM) and the stepwise mutation model (SMM)) provide some guides to predictions from which the possible effect of population substructuring may be evaluated, assuming that the subpopulations do not exchange any genes among them during evolution. The theory predicts that the loci with larger mutation rate, and consequently showing greater heterozygosity within subpopulations, should exhibit a smaller proportional heterozygote deficiency (GST) and, hence, the effects of population substructuring should be minimal at the hypervariable DNA loci (an order of magnitude smaller than that at the blood group and protein loci). Applications of this theory to data on six Variable Number of Tandem Repeat (VNTR) loci and five short tandem repeat (STR) loci in the major cosmopolitan populations of the USA show that while the VNTR loci often exhibit a large significant heterozygote deficiency, the STR loci do not show a similar tendency. This discordant finding may be ascribed to the limitations, coalescence and nondetectability of alleles associated with the restriction fragment length polymorphism (RFLP) analysis through which the VNTR loci are scored. Such limitations do not apply to the polymerase chain reaction (PCR) method, through which the STR loci are scored. The implications of these results are discussed in the context of the forensic use of DNA typing data.

Population genetics of dinucleotide (dC-dA)n.(dG-dT)n polymorphisms in world populations

We have characterized eight dinucleotide (dC-dA)n.(dG-dT)n repeat loci located on human chromosome 13q in eight human populations and in a sample of chimpanzees. Even though there is substantial variation in allele frequencies at each locus, at a given locus the most frequent alleles are shared by all human populations. The level of heterozygosity is reduced in isolated or small populations, such as the Pehuenche Indians of Chile, the Dogrib of Canada, and the New Guinea highlanders. On the other hand, larger average heterozygosities are observed in large and cosmopolitan populations, such as the Sokoto population from Nigeria and German Caucasians. Conformity with Hardy-Weinberg equilibrium is generally observed at these loci, unless (a) a population is isolated or small or (b) the repeat motif of the locus is not perfect (e.g., D13S197). Multilocus genotype probabilities at these microsatellite loci do not show departure from the independence rule, unless the loci are closely linked. The allele size distributions at these (CA)n loci do not follow a strict single-step stepwise-mutation model. However, this features does not compromise the ability to detect population affinities, when these loci are used simultaneously. The microsatellite loci examined here are present and, with the exception of the locus D13S197, are polymorphic in the chimpanzees, showing an overlapping distribution of allele sizes with those observed in human populations.

D1S80 population data in African Americans, Caucasians, southeastern Hispanics, southwestern Hispanics, and Orientals

Allele frequencies for the locus D1S80 were determined in African American, Caucasian, Southeastern Hispanic, Southwestern Hispanic, and Oriental sample populations using the polymerase chain reaction and subsequent electrophoresis and silver staining of the amplified products. Due to the presence of anodal and cathodal electrophoretic variants (in reference to the steps in an allelic ladder), allele frequencies were established using a classification protocol based on the steps in the allelic ladder. All sample populations met Hardy-Weinberg expectations for D1S80. In addition, there was no evidence for association of alleles between the loci D1S80 and D1S7. The product of allele frequencies from the data from the sample populations in this study can be used in forensic analyses and paternity tests to estimate the frequency of a D1S80 DNA genotype.

Phylogeny of mitochondrial DNA clones in tassel-eared squirrels Sciurus aberti

The tassel-eared squirrel, Sciurus aberti, includes six subspecies which occupy restrictive and apparently identical habitats in Ponderosa pine forests in the south-western United States and Mexico; the strict habitat requirement of this species is based on dietary requirements which are only fulfilled in these forests. To examine evolutionary relationships among certain subspecies of S. aberti, we obtained estimates of nucleotide diversity within subspecies as well as nucleotide divergence between subspecies using mitochondrial DNA (mtDNA) analysis. Restriction site polymorphisms were identified in samples of the four US subspecies: S. a. aberti (Abert), S. a. kaibabensis (Kaibab), S. a. ferreus (Ferreus), and S. a. chuscensis (Chuska) Fourteen mtDNA clones were resolved that were, with one exception, uniquely subspecific. Dendrograms constructed by neighbour-joining and maximum parsimony methods revealed two major assemblages: (1) an Abert/Kaibab group; and (2) a Ferreus/Chuska group. The Abert vs. Ferreus clones exhibited the greatest net nucleotide divergence, with a lineage separation estimate approximating 572,000 years ago assuming a nucleotide substitution rate of 7.15 x 10(-9)/year/site. Five out of ten Chuska squirrels shared a clone with one Abert sample; the relative sizes of these two populations and their respective ranges as well as their close proximity support the proposal for relatively recent intermixing of Abert and Chuska populations resulting in what appears to be Abert-->Chuska migration. Nucleotide diversity within subspecies ranked as Kaibab < Ferreus < Abert < Chuska; the relatively high diversity for the Chuska sample is based on the apparent introgression of Abert mtDNA. The relative diversity exhibited by Kaibab, Ferreus and Aberti samples corresponds to the range size of the respective subspecies.

Estimating the frequency of nonpaternity in Switzerland

In linkage analysis and gene mapping in general, typing error and/or errors in specifying pedigree relationships between individuals typically lead to an upwards bias in recombination fraction estimates and to a loss of power for detecting linkage. [Ott J: Analysis of Human Genetic Linkage. Baltimore, Johns Hopkins University Press, 1991]. The presence of nonpaternity is the most common source of error of misspecification of pedigree relationships between individuals that can have a negative impact on pedigree analysis. We have developed estimates of underlying nonpaternity probabilities based on observed Mendelian inconsistencies. The methods were applied to a sample of 1,607 children and their parents. Genetic marker data were investigated to test for exclusions due to nonpaternity. Among the 1,607 children, 11 or 0.68% exclusions were found. When a constant nonpaternity rate was assumed for each child, its maximum likelihood estimate turned out to be 0.78% with asymptotic 95% confidence limits of 0.41% and 1.35%. When varying nonpaternity rates were assumed, its mean was estimated as 0.83% with asymptotic 95% confidence limits of 0.32% and 1.33%.

Genetic characterization of American and Western Samoans

The Samoan islands were politically separated into American Samoa and Western Samoa in the early 1900s. Economic modernization is far more extensive in American Samoa. However, the Samoan archipelago has maintained a remarkable degree of sociocultural homogeneity, including intermarriage. The sociocultural exchanges presumably led to genetic homogeneity between the two Samoas. Detailed genetic comparisons and characterizations of Samoans are scanty, however. As part of a multidisciplinary study of modernization and cardiovascular risk factors in adults, we analyzed nine hypervariable nuclear DNA (HVR) and four serum protein polymorphisms in the two Samoan groups. The average heterozygosities at both DNA and serum protein loci are comparable in the two groups. As expected, the HVR loci reveal a high degree of variability (heterozygosity 30-87%) compared with the serum protein loci (heterozygosity 1-52%). A large proportion of alleles at the HVR loci, ranging from 50% to 100%, are shared between American and Western Samoa. With the exceptions of the D1S80 locus in American Samoa and the D13S118 locus in Western Samoa, the genotype distributions at all loci conform to their respective Hardy-Weinberg expectations. Sporadic occurrence of the F13B*2 allele at the F13B locus in Samoans indicates a low level of European admixture because this allele is unique to Europeans. The calculated zero values of kinship coefficients and standard genetic distances indicate minimal population differentiation between the two Samoan groups.

Population genetic characteristics of the D1S80 locus in seven human populations

We have analyzed the allele frequency distribution at the highly polymorphic variable number of tandem repeat (VNTR) locus D1S80 (pMCT118) in seven ethnic populations (namely, New Guinea Highlanders of Papua New Guinea, Dogrib Indians of Canada, Pehuenche Indians of Chile, American and Western Samoans, Kacharis of Northeast India, and German Caucasians) using the polymerase chain reaction (PCR) technique. In the pooled sample of 443 unrelated individuals 20 segregating alleles were detected. A trimodal pattern of allelic distribution is present in the majority of populations and is indicative of the evolutionary antiquity of the polymorphism at this locus. In spite of the observed high degree of polymorphism (expected heterozygosity 56%-86%), with a single exception--the marginally significant P value (0.04) of the exact test in American Samoans--the genotype distributions in all populations conform to their respective Hardy-Weinberg expectations. Summary statistics indicate that, in general, the allele frequency distribution at this locus may be approximated by the infinite allele model. The data also demonstrate that alleles that are shared by all populations have the highest average frequency within populations. Furthermore, the kinship bioassay analysis demonstrates that the extensive variation observed at the D1S80 locus is at the interindividual within population level, which dwarfs any interpopulation allele frequency variation, consistent with the population dynamics of hypervariable polymorphisms. These characteristics of the D1S80 locus make it a very useful marker for population genetic research, genetic linkage studies, forensic identification of individuals, and for determination of biological relatedness of individuals.

Kin selection, social structure, gene flow, and the evolution of chimpanzees

Hypotheses about chimpanzee social behavior, phylogeography, and evolution were evaluated by noninvasive genotyping of free-ranging individuals from 20 African sites. Degrees of relatedness among individuals in one community were inferred from allele-sharing at eight nuclear simple sequence repeat (SSR) loci. Males are related on the order of half-siblings, and homozygosity is significantly increased at several SSR loci compared to Hardy-Weinberg expectations. These data support the kin-selection hypothesis for the evolution of cooperation among males. Sequence variation patterns at two mitochondrial loci indicate historically high long-distance gene flow and clarify the relationships among three allopatric subspecies. The unexpectedly large genetic distance between the western subspecies, Pan troglodytes verus, and the other two subspecies suggests a divergence time of about 1.58 million years. This result, if confirmed at nuclear loci and supported by eco-behavioral data, implies that P. t. verus should be elevated to full species rank.

Nondetectability of restriction fragments and independence of DNA fragment sizes within and between loci in RFLP typing of DNA

We provide experimental evidence showing that, during the restriction-enzyme digestion of DNA samples, some of the HaeIII-digested DNA fragments are small enough to prevent their reliable sizing on a Southern gel. As a result of such nondetectability of DNA fragments, individuals who show a single-band DNA profile at a VNTR locus may not necessarily be true homozygotes. In a population database, when the presence of such nondetectable alleles is ignored, we show that a pseudodependence of alleles within as well as across loci may occur. Using a known statistical method, under the hypothesis of independence of alleles within loci, we derive an efficient estimate of null allele frequency, which may be subsequently used for testing allelic independence within and across loci. The estimates of null allele frequencies, thus derived, are shown to agree with direct experimental data on the frequencies of HaeIII-null alleles. Incorporation of null alleles into the analysis of the forensic VNTR database suggests that the assumptions of allelic independence within and between loci are appropriate. In contrast, a failure to incorporate the occurrence of null alleles would provide a wrong inference regarding the independence of alleles within and between loci.

Conservation of human chromosome 13 polymorphic microsatellite (CA)n repeats in chimpanzees

Tandemly repeated (dC-dA)n.(dG-dT)n sequences occur abundantly and are found in most eukaryotic genomes. To investigate the level of conservation of these repeat sequences in nonhuman primates, we have analyzed seven human chromosome 13 dinucleotide (CA)n repeat loci in chimpanzees by DNA amplification using primers designed for analysis of human loci. Comparable levels of polymorphism at these loci in the two species, revealed by the number of alleles, heterozygosity, and allele sizes, suggest that the (CA)n repeat arrays and their genomic locations are highly conserved. Even though the proportion of shared alleles between the two species varies enormously and the modal alleles are not the same, allelic lengths at each locus in the chimpanzees are detected within the bounds of the allele size range observed in humans. A similar observation has been noted in a limited number of gorillas and orangutans. Using a new measure of genetic distance that takes into account the size of alleles, we have compared the genetic distance between humans and chimpanzees. The genetic distance between these two species was found to be ninefold smaller than expected assuming there is no selection or mutational bias toward retention of (CA)n repeat arrays. These findings suggest a functional significance for these microsatellite loci.

Evaluation of 13 short tandem repeat loci for use in personal identification applications

Personal identification by using DNA typing methodologies has been an issue in the popular and scientific press for several years. We present a PCR-based DNA-typing method using 13 unlinked short tandem repeat (STR) loci. Validation of the loci and methodology has been performed to meet standards set by the forensic community and the accrediting organization for parentage testing. Extensive statistical analysis has addressed the issues surrounding the presentation of "match" statistics. We have found STR loci to provide a rapid, sensitive, and reliable method of DNA typing for parentage testing, forensic identification, and medical diagnostics. Valid statistical analysis is generally simpler than similar analysis of RFLP-VNTR results and provides powerful statistical evidence of the low frequency of random multilocus genotype matching.

Linkage disequilibria among (CA)n polymorphisms in the human dystrophin gene and their implications in carrier detection and prenatal diagnosis in Duchenne and Becker muscular dystrophies

Four short tandem repeat loci, characterized by length polymorphisms of (CA)n repeats, have been detected within introns 44, 45, 49, and 50 of the human dystrophin gene. The predicted heterozygosities for these loci range from 72 to 93%, and observed allele numbers range from 6 to 19 in 57 normal chromosomes, revealing their high degree of polymorphism. Evidence for significant disequilibria between the loci within introns 49 and 50 is found. These data appear to be consistent with observations of recombination frequencies between these markers and the length of the intron 44 in relation to the entire region. In addition, these four loci are collectively found to be 100% informative in carrier detection/prenatal diagnosis of Becker and Duchenne muscular dystrophies (B/DMD), whereas scoring the (CA)n markers within introns 45 and 49 alone gives a 99.6% success rate.

Paternity testing in the DNA era

Paternity can be efficiently and effectively tested using multi-allelic variable number of tandem repeat (VNTR) polymorphisms. This can be achieved by using multilocus DNA fingerprinting probes or batteries of single-locus minisatellite probes, or by PCR testing of sets of minisatellite and microsatellite loci. Although some problems of statistics and population genetics remain to be overcome, the informativeness of these hypervariable loci is so high that paternity disputes can be resolved with certainty in virtually every case.