The East African coast has been a nexus of trade and cultural contact in the Indian Ocean basin for at least the last 2,000 years. Unlike their neighbors, the Swahili residents of the East African coast are predominantly Muslim, wrote their language in the Arabic script prior to European colonization, and have towns and cities with stone architecture. Until the 1970s, both foreign and many local commentators generally viewed these "civilized" features of the Swahili as introduced by trade and intermarriage with Arab and Persian traders. In addition, Swahili oral and written histories do little to dispel this notion of external influence, with traditions tracing family lineages to foreign ancestors. Archeological and linguistic research has significantly changed this early view of Swahili origins. It is now generally accepted that the Swahili are primarily an African population speaking a Bantu-derived language whose cultural history derives much more from in situ indigenous development than external influences.

In our interdisciplinary research, my collaborators and I aim to determine the genetic relationships between ancient and modern Swahili and the origins of precolonial East African urban residents from genetic and skeletal evidence, and to secure accelerated radiocarbon dates necessary for determining the chronology of urbanism. These data will elucidate the ethnic and biological genealogy of Swahili populations and contribute to the resolution of long-standing questions in the archaeology of urbanism in East Africa including: (1) what were the cultural, technological, and biological relationships between urban and rural populations of the coast; (2) when and in what ways did East Africa become part of Indian Ocean and Persian Gulf trading systems; and (3) what forms did this engagement take and how did they impact regional and inter-regional interaction spheres?

On the basis of relatively limited mitochondrial data (mtDNA – a maternal history marker), it has been hypothesized that the contemporary Yemeni population includes a significant African heritage that derives primarily from the entry of African women into the population within the last 2000 years. In order to test this hypothesis, this project proposes a more thorough genetic analysis (via autosomal genotyping) of these samples to complement the basic population genetic data already collected. In the parent study we have already determined mitochondrial DNA haplotypes that we will use to infer maternal history, Y chromosome haplotypes that we will use to infer paternal history, and a preliminary set of autosomal (the set of nuclear chromosomes excluding X and Y) genotypes that we have begun to use to infer admixture proportions and the demographic history of this population. However, the autosomal data collected so far have limited statistical power. We have collected high-resolution SNP genotype data and with the increased power achievable with these markers, we will be able to better evaluate the timing and amount of African admixture in this population.

Yemen has been a nexus of trade for at least the last 2000 years. Published research and our preliminary results reveal a population primarily of Middle Eastern origin, but with significant African and southeast Asian admixture. As such, our sample is representative of a population at the interface of at least three large continental populations. Multiple lines of evidence now suggest that these regional or continental populations have developed adaptations to local conditions or behaviors, some of which may be globally advantageous. One example of this phenomenon is lactase persistence – the retention of the ability to digest milk in adults. It is now known that this ability developed separately in northern Europe and in cattle-rearing populations of east Africa. Using high resolution SNP genotypes from microarray technologies, we will examine the relative prevalence of these alternative alleles (and other adaptations) in the contemporary Yemeni population in order to infer the rate and direction of the gene flow of adaptive alleles.

Social groups in Yemen are ordered on the basis of paternal descent lines in a hierarchy of social rank. It has been hypothesized that these paternal lineages may be more reflective of past political and social alliance than actual genetic relations. Each village in Yemen consists of one or more of these paternal lineages. A small village may have only one lineage while a large village may have five or more. We collected samples from most of the paternal lineages in a very large village (census population of greater than 6000 individuals). We have collected data from Y chromosome microsatellite markers from these individuals to assess the correspondance between the social and genetic genealogies.