Several decades of research have convincingly shown that classical human leukocyte antigen (HLA) loci bear signatures of natural selection. Despite this conclusion, many questions remain regarding the type of selective regime acting on these loci, the time frame at which selection acts, and the functional connections between genetic variability and natural selection. In this review, we argue that genomic datasets, in particular those generated by next-generation sequencing (NGS) at the population scale, are transforming our understanding of HLA evolution. We show that genomewide data can be used to perform robust and powerful tests for selection, capable of identifying both positive and balancing selection at HLA genes. Importantly, these tests have shown that natural selection can be identified at both recent and ancient timescales. We discuss how findings from genomewide association studies impact the evolutionary study of HLA genes, and how genomic data can be used to survey adaptive change involving interaction at multiple loci. We discuss the methodological developments which are necessary to correctly interpret genomic analyses involving the HLA region. These developments include adapting the NGS analysis framework so as to deal with the highly polymorphic HLA data, as well as developing tools and theory to search for signatures of selection, quantify differentiation, and measure admixture within the HLA region. Finally, we show that high throughput analysis of molecular phenotypes for HLA genes—namely transcription levels—is now a feasible approach and can add another dimension to the study of genetic variation.

Next-generation sequencing (NGS) technologies have become the standard for data generation in studies of population genomics, as the 1000 Genomes Project (1000G). However, these techniques are known to be problematic when applied to highly polymorphic genomic regions, such as the human leukocyte antigen (HLA) genes. Because accurate genotype calls and allele frequency estimations are crucial to population genomics analyses, it is important to assess the reliability of NGS data. Here, we evaluate the reliability of genotype calls and allele frequency estimates of the single-nucleotide polymorphisms (SNPs) reported by 1000G (phase I) at five HLA genes (HLA-A, -B, -C, -DRB1, and -DQB1). We take advantage of the availability of HLA Sanger sequencing of 930 of the 1092 1000G samples and use this as a gold standard to benchmark the 1000G data. We document that 18.6% of SNP genotype calls in HLA genes are incorrect and that allele frequencies are estimated with an error greater than ±0.1 at approximately 25% of the SNPs in HLA genes. We found a bias toward overestimation of reference allele frequency for the 1000G data, indicating mapping bias is an important cause of error in frequency estimation in this dataset. We provide a list of sites that have poor allele frequency estimates and discuss the outcomes of including those sites in different kinds of analyses. Because the HLA region is the most polymorphic in the human genome, our results provide insights into the challenges of using of NGS data at other genomic regions of high diversity.

When humans moved from Asia toward the Americas over 18,000 y ago and eventually peopled the New World they encountered a new environment with extreme climate conditions and distinct dietary resources. These environmental and dietary pressures may have led to instances of genetic adaptation with the potential to influence the phenotypic variation in extant Native American populations. An example of such an event is the evolution of the fatty acid desaturases (FADS) genes, which have been claimed to harbor signals of positive selection in Inuit populations due to adaptation to the cold Greenland Arctic climate and to a protein-rich diet. Because there was evidence of intercontinental variation in this genetic region, with indications of positive selection for its variants, we decided to compare the Inuit findings with other Native American data. Here, we use several lines of evidence to show that the signal of FADS-positive selection is not restricted to the Arctic but instead is broadly observed throughout the Americas. The shared signature of selection among populations living in such a diverse range of environments is likely due to a single and strong instance of local adaptation that took place in the common ancestral population before their entrance into the New World. These first Americans peopled the whole continent and spread this adaptive variant across a diverse set of environments.

As whole-genome sequencing (WGS) becomes the gold standard tool for studying population genomics and medical applications, data on diverse non-European and admixed individuals are still scarce. Here, we present a high-coverage WGS dataset of 1,171 highly admixed elderly Brazilians from a census-based cohort, providing over 76 million variants, of which ~2 million are absent from large public databases. WGS enables identification of ~2,000 previously undescribed mobile element insertions without previous description, nearly 5 Mb of genomic segments absent from the human genome reference, and over 140 alleles from HLA genes absent from public resources. We reclassify and curate pathogenicity assertions for nearly four hundred variants in genes associated with dominantly-inherited Mendelian disorders and calculate the incidence for selected recessive disorders, demonstrating the clinical usefulness of the present study. Finally, we observe that whole-genome and HLA imputation could be significantly improved compared to available datasets since rare variation represents the largest proportion of input from WGS. These results demonstrate that even smaller sample sizes of underrepresented populations bring relevant data for genomic studies, especially when exploring analyses allowed only by WGS.

Next-generation sequencing (NGS) technologies have become the standard for data generation in studies of population genomics, as the 1000 Genomes Project (1000G). However, these techniques are known to be problematic when applied to highly polymorphic genomic regions, such as the human leukocyte antigen (HLA) genes. Because accurate genotype calls and allele frequency estimations are crucial to population genomics analyses, it is important to assess the reliability of NGS data. Here, we evaluate the reliability of genotype calls and allele frequency estimates of the single-nucleotide polymorphisms (SNPs) reported by 1000G (phase I) at five HLA genes (HLA-A, -B, -C, -DRB1, and -DQB1). We take advantage of the availability of HLA Sanger sequencing of 930 of the 1092 1000G samples and use this as a gold standard to benchmark the 1000G data. We document that 18.6% of SNP genotype calls in HLA genes are incorrect and that allele frequencies are estimated with an error greater than 60.1 at approximately 25% of the SNPs in HLA genes. We found a bias toward overestimation of reference allele frequency for the 1000G data, indicating mapping bias is an important cause of error in frequency estimation in this dataset. We provide a list of sites that have poor allele frequency estimates and discuss the outcomes of including those sites in different kinds of analyses. Because the HLA region is the most polymorphic in the human genome, our results provide insights into the challenges of using of NGS data at other genomic regions of high diversity. KEYWORDS