Detection of differential gene flow from patterns of quantitative variation.
ABSTRACT A major goal in anthropological genetics is the assessment of the effects of different microevolutionary forces. Harpending and Ward (1982) developed a model that aids in this effort by comparing observed and expected heterozygosity within populations in a local region. The expected heterozygosity within a population is a function of the total heterozygosity of the entire region and the distance of the population from the regional mean centroid of allele frequencies. Greater than average gene flow from an external source will result in a population having greater heterozygosity than expected. Less than average gene flow from an external source will result in a population having less heterozygosity than expected. We extend the Harpending-Ward model to quantitative traits using an equal and additive effects model of inheritance. Here the additive genetic variance within a population is directly proportional to heterozygosity, and its expectation is directly proportional to the genetic distance from the centroid. Under certain assumptions the expectations for phenotypic variances are similar. Observed and expected genetic or phenotypic variance can then be compared to assess the effects of differential external gene flow. When the additive genetic covariance matrix or heritabilities are not known, the phenotypic covariance matrix can be used to provide a conservative application of the model. In addition, we develop new methods for estimation of the genetic relationship matrix (R) from quantitative traits. We apply these models to two data sets: (1) six principal components derived from twenty dermatoglyphic ridge count measures for nine villages in Nepal and (2) ten anthropometric measurements for seven isolated populations in western Ireland. In both cases both the univariate and multivariate analyses provide results that can be directly interpreted in terms of historically known patterns of gene flow.
SourceAvailable from: Nuria Torrescano-ValleArchaeology and Bioarchaeology of Population Movement among the Prehispanic Maya, 1a. edited by Andrea Cucina, 01/2015: chapter Chapter 4 Calakmul: Power, Perseverance, and Persistence: pages 37-50; Springer., ISBN: ISBN 978-3-319-10857-5 ISBN 978-3-319-10858-2 (eBook)
[Show abstract] [Hide abstract]
ABSTRACT: Genetic resemblances among groups are non-randomly distributed in humans. This population structure may influence the correlations between traits and environmental drivers of natural selection thus complicating the interpretation of the fossil record when modern human variation is used as a referential model. In this paper, we examine the effects of population structure and natural selection on postcranial traits that reflect body size and shape with application to the more general issue of how climate – using latitude as a proxy – has influenced hominin morphological variation. We compare models that include terms reflecting population structure, ascertained from globally distributed microsatellite data, and latitude on postcranial phenotypes derived from skeletal dimensions taken from a large global sample of modern humans. We find that models with a population structure term fit better than a model of natural selection along a latitudinal cline in all cases. A model including both latitude and population structure terms is a good fit to distal limb element lengths and bi-iliac breadth, indicating that multiple evolutionary forces shaped these morphologies. In contrast, a model that included only a population structure term best explained femoral head diameter and the crural index. The results demonstrate that population structure is an important part of human postcranial variation, and that clinally distributed natural selection is not sufficient to explain among-group differentiation. The distribution of human body form is strongly influenced by the contingencies of modern human origins, which calls for new ways to approach problems in the evolution of human variation, past and present.Journal of Human Evolution 11/2014; DOI:10.1016/j.jhevol.2014.07.006 · 3.87 Impact Factor
The South African Archaeological Bulletin 01/2012; 67:44-51.