Resumen de Human population genetics in the Mediterranean region. From single markers to whole-genome sequencing
The human populations of the Mediterranean region have been thoroughly studied. Nonetheless, there are numerous aspects of their demographic and epidemiological history that remain uncertain to date. The growing use of high-throughput genetic data is helping to unravel these questions. A crucial part in designing a population genetics study is to determine the type and amount of genetic data that is needed in order to obtain accurate results. In the case of analyses that deal with large-scale population processes, such as continental barriers, or that focus on some specific candidate genes, sparse genetic data with a high degree of variation could be an option that allows for larger sample sizes. However, to obtain more refined results, such as estimating the date of an admixture event, inferring historical changes in population sizes, or focusing on restricted geographic areas, it is required the use of array-based or, ideally, whole-genome sequencing data. In this doctoral thesis, I present four studies in which the density of genetic markers employed vary in accordance to the depth of the analysis. Globally, this work contributes to the knowledge in the sphere of the population genetics of Mediterranean populations. The first survey supports the role of the Mediterranean sea as a historical barrier to gene flow, making use of three single nucleotide polymorphisms (SNP) located within or around LIN28B, a gene associated with cancer. One of these markers, rs221639, shows a specially high degree of variation between Mediterranean populations. The second article analyzes the presence of a sub-Saharan genetic component in four CAD-associated genomic regions in Mediterranean populations. This component is more prevalent in the North African coast, suggesting a more intense sub-Saharan gene flow than in Southern Europe. Furthermore, D-statistics suggest potential sub-Saharan introgression at the 10q11 region, which includes CXCL12, a gene that codes for a chemokine ligand linked to cardiovascular disease with protective effects. In the third study, the level of resolution achieved by array-based genome-wide data allows to refine previous estimates of a Sephardic component present in current Iberian populations. The incorporation of neighbouring populations to the analyses shows a gene flow process from the Iberian Peninsula outwards, which could be reflecting a migration pattern followed by the expelled Sephardic Jews. Finally, I present an analysis of deep-coverage whole-genome sequencing data from the Spanish Eastern Pyrenean population. In combination with accurate genealogical information of the samples, the quality of the data allows to analyze the demographic history of the Spanish Eastern Pyreneans with a high degree of detail using, among others, haplotype-based and deep learning methods. Namely, the Spanish Eastern Pyreneans appear as a distinct group within the Iberian populations, closely related to the Basques. In addition, this human group presents fine-scale population structure, and it has undergone a historical isolation process involving a reduction of the effective size, whose epidemiological consequence has been a significant depletion of many rare and highly deleterious mutations.
