Neither self-reported ethnicity nor declared family origin are reliable indicators of genomic ancestry

Abstract

Ancestry information can be useful in investigations of diseases with a genetic or infectious background. As the Brazilian population is highly admixed physical traits tend to be poor indicators of ancestry. The assessment of ancestry by ancestry informative markers (AIMs) can exclude the subjectivity of self-declared ethnicity and reported family origin. We aimed to evaluate the reliability of self-reported ethnicity or reported family origin as indicators of genomic ancestry in a female population from the Southeast of Brazil. Two cohorts were included: 404 women asked to self-report their ethnicity (Pop1) and 234 women asked to report their family’s origin (Pop2). Identification of AIMs was performed using a panel of 61 markers and results were plotted against parental populations—Amerindian, Western European and Sub-Saharan African—using Structure v2.3.4. In Pop1 57.4 % of women self-reported as white, 34.6 % as brown and 8.0 % as black. Median global European, Amerindian and African contributions were 66.8, 12.6 and 16.6 %. In Pop2, 66.4 % of women declared European origin, 23.9 % African origin and 26.9 % Amerindian. Median global European, Amerindian and African contributions were 80.8, 7.3 and 7.6 %, respectively. Only 31.0 and 21.0 % of the global variation in African and European contributions, respectively, could be explained by self-reported ethnicity and reported family origin only accounted for 20.0 and 5.0 % of the variations observed in African and European ancestries, respectively. Amerindian ancestry did not influence self-reported ethnicity or declared family origin. Neither self-reported ethnicity nor declared family origin are reliable indicators of genomic ancestry in these Brazilian populations.