ABSTRACT
Here we evaluate the accuracy of prediction for eye, hair and skin pigmentation in a dataset of > 6500 individuals from Mexico, Colombia, Peru, Chile and Brazil (including genome-wide SNP data and quantitative/categorical pigmentation phenotypes - the CANDELA dataset CAN). We evaluated accuracy in relation to different analytical methods and various phenotypic predictors. As expected from statistical principles, we observe that quantitative traits are more sensitive to changes in the prediction models than categorical traits. We find that Random Forest or Linear Regression are generally the best performing methods. We also compare the prediction accuracy of SNP sets defined in the CAN dataset (including 56, 101 and 120 SNPs for eye, hair and skin colour prediction, respectively) to the well-established HIrisPlex-S SNP set (including 6, 22 and 36 SNPs for eye, hair and skin colour prediction respectively). When training prediction models on the CAN data, we observe remarkably similar performances for HIrisPlex-S and the larger CAN SNP sets for the prediction of hair (categorical) and eye (both categorical and quantitative), while the CAN sets outperform HIrisPlex-S for quantitative, but not for categorical skin pigmentation prediction. The performance of HIrisPlex-S, when models are trained in a world-wide sample (although consisting of 80% Europeans, https://hirisplex.erasmusmc.nl), is lower relative to training in the CAN data (particularly for hair and skin colour). Altogether, our observations are consistent with common variation of eye and hair colour having a relatively simple genetic architecture, which is well captured by HIrisPlex-S, even in admixed Latin Americans (with partial European ancestry). By contrast, since skin pigmentation is a more polygenic trait, accuracy is more sensitive to prediction SNP set size, although here this effect was only apparent for a quantitative measure of skin pigmentation. Our results support the use of HIrisPlex-S in the prediction of categorical pigmentation traits for forensic purposes in Latin America, while illustrating the impact of training datasets on its accuracy.
Subject(s)
Eye Color/genetics , Hair Color/genetics , Polymorphism, Single Nucleotide , Skin Pigmentation/genetics , Datasets as Topic , Genetics, Population , Genotype , Humans , Latin America , Logistic Models , PhenotypeABSTRACT
[This corrects the article DOI: 10.1371/journal.pone.0169287.].
ABSTRACT
The expression of facial asymmetries has been recurrently related with poverty and/or disadvantaged socioeconomic status. Departing from the developmental instability theory, previous approaches attempted to test the statistical relationship between the stress experienced by individuals grown in poor conditions and an increase in facial and corporal asymmetry. Here we aim to further evaluate such hypothesis on a large sample of admixed Latin Americans individuals by exploring if low socioeconomic status individuals tend to exhibit greater facial fluctuating asymmetry values. To do so, we implement Procrustes analysis of variance and Hierarchical Linear Modelling (HLM) to estimate potential associations between facial fluctuating asymmetry values and socioeconomic status. We report significant relationships between facial fluctuating asymmetry values and age, sex, and genetic ancestry, while socioeconomic status failed to exhibit any strong statistical relationship with facial asymmetry. These results are persistent after the effect of heterozygosity (a proxy for genetic ancestry) is controlled in the model. Our results indicate that, at least on the studied sample, there is no relationship between socioeconomic stress (as intended as low socioeconomic status) and facial asymmetries.
Subject(s)
Facial Asymmetry/epidemiology , Facial Asymmetry/genetics , Adolescent , Adult , Female , Heterozygote , Hispanic or Latino/genetics , Humans , Male , Middle Aged , Social Class , Young AdultABSTRACT
Fluctuating and directional asymmetry are aspects of morphological variation widely used to infer environmental and genetic factors affecting facial phenotypes. However, the genetic basis and environmental determinants of both asymmetry types is far from being completely known. The analysis of facial asymmetries in admixed individuals can be of help to characterize the impact of a genome's heterozygosity on the developmental basis of both fluctuating and directional asymmetries. Here we characterize the association between genetic ancestry and individual asymmetry on a sample of Latin-American admixed populations. To do so, three-dimensional (3D) facial shape attributes were explored on a sample of 4,104 volunteers aged between 18 and 85 years. Individual ancestry and heterozygosity was estimated using more than 730,000 genome-wide markers. Multivariate techniques applied to geometric morphometric data were used to evaluate the magnitude and significance of directional and fluctuating asymmetry (FA), as well as correlations and multiple regressions aimed to estimate the relationship between facial FA scores and heterozygosity and a set of covariates. Results indicate that directional and FA are both significant, the former being the strongest expression of asymmetry in this sample. In addition, our analyses suggest that there are some specific patterns of facial asymmetries characterizing the different ancestry groups. Finally, we find that more heterozygous individuals exhibit lower levels of asymmetry. Our results highlight the importance of including ancestry-admixture estimators, especially when the analyses are aimed to compare levels of asymmetries on groups differing on socioeconomic levels, as a proxy to estimate developmental noise.