Ancestralidade e casamentos preferenciais em populações brasileiras

Abstract

Human populations shows a trend of marriages between individuals with similar socioeconomic status, educational level and genomic ancestry. This behavior violates the assumption of random marriages of several statistical models in genetics populations and medical genetics, which can lead to misleading results in quantitative genetics analysis and demographic inferences. Therefore, the knowledge about the trends of marriages in Brasil is crucial to the progress of Brazilian populationgenomic studies. In previous studies our group verified the occurrence of ancestry-assortment for the three Brazilian populations of the EPIGEN-Brasil (The Latin American initiative in population genomics and genetic epidemiology): Salvador BA, Pelotas RS, Bambuí MG. In this project we investigated the patterns of ancestry-assortment for different socioeconomic status and educational levels in these populations. The 6.487 samples from the three population-based cohorts were genotypedfor 2.5 million SNPs (single nucleotide polymorphism) and each cohort was stratified in categories based on individual educational level or income. Homozygosity excess in SNPs that are highly differentiated between ancestral populations indicate ancestry-assortment, therefore we estimated the Spearman's rank correlation (rho) for the ancestry informativeness and the homozygosity excess estimated for each SNP. Our study finds that the ancestry-assortment is significantly affected byeducational level and income. Salvador and Pelotas showed a crescent pattern with the most remarkable correlation in the high educational level compared to the low and middle levels. The opposite is seen in Bambuí, that shows a more expressive evidence of ancestry-assortment at the lower educational level compared to the middle and high levels. Thus, our results enables that the Brazilian population genomic studies consider those regional differences in the assortative mating.