Implications of the Admixture Process in Skin Color Molecular Assessment

Caio Cesar Silva de Cerqueira, Tábita Hünemeier, Jorge Gomez-Valdés, Virgínia Ramallo, Carla Daiana Volasko-Krause, Ana Angélica Leal Barbosa, Pedro Vargas-Pinilla, Rodrigo Ciconet Dornelles, Danaê Longo, Francisco Rothhammer, Gabriel Bedoya, Samuel Canizales-Quinteros, Victor Acuña-Alonzo, Carla Gallo, Giovanni Poletti, Rolando González-José, Francisco Mauro Salzano, Sídia Maria Callegari-Jacques, Lavínia Schuler-Faccini, Andrés Ruiz-Linares, Maria Cátira Bortolini

The understanding of the complex genotype-phenotype architecture of human pigmentation has clear implications for the evolutionary history of humans, as well as for medical and forensic practices. Although dozens of genes have previously been associated with human skin color, knowledge about this trait remains incomplete. In particular, studies focusing on populations outside the European-North American axis are rare, and, until now, admixed populations have seldom been considered. The present study was designed to help fill this gap. Our objective was to evaluate possible associations of 18 single nucleotide polymorphisms (SNPs), located within nine genes, and one pseudogene with the Melanin Index (MI) in two admixed Brazilian populations (Gaucho, N = 352; Baiano, N = 148) with different histories of geographic and ethnic colonization. Of the total sample, four markers were found to be significantly associated with skin color, but only two (SLC24A5 rs1426654, and SLC45A2 rs16891982) were consistently associated with MI in both samples (Gaucho and Baiano). Therefore, only these 2 SNPs should be preliminarily considered to have forensic significance because they consistently showed the association independently of the admixture level of the populations studied. We do not discard that the other two markers (HERC2 rs1129038 and TYR rs1126809) might be also relevant to admixed samples, but additional studies are necessary to confirm the real importance of these markers for skin pigmentation. Finally, our study shows associations of some SNPs with MI in a modern Brazilian admixed sample, with possible applications in forensic genetics. Some classical genetic markers in Euro-North American populations are not associated with MI in our sample. Our results point out the relevance of considering population differences in selecting an appropriate set of SNPs as phenotype predictors in forensic practice.