Searching for improvements in predicting human eye colour from DNA

Abstract

Increasing understanding of human genome variability allows for better use of the predictive potential of DNA. An obvious direct application is the prediction of the physical phenotypes. Significant success has been achieved, especially in predicting pigmentation characteristics, but the inference of some phenotypes is still challenging. In search of further improvements in predicting human eye colour, we conducted whole-exome (enriched in regulome) sequencing of 150 Polish samples to discover new markers. For this, we adopted quantitative characterization of eye colour phenotypes using high-resolution photographic images of the iris in combination with DIAT software analysis. An independent set of 849 samples was used for subsequent predictive modelling. Newly identified candidates and 114 additional literature-based selected SNPs, previously associated with pigmentation, and advanced machine learning algorithms were used. Whole-exome sequencing analysis found 27 previously unreported candidate SNP markers for eye colour. The highest overall prediction accuracies were achieved with LASSO-regularized and BIC-based selected regression models. A new candidate variant, rs2253104, located in the ARFIP2 gene and identified with the HyperLasso method, revealed predictive potential and was included in the best-performing regression models. Advanced machine learning approaches showed a significant increase in sensitivity of intermediate eye colour prediction (up to 39%) compared to 0% obtained for the original IrisPlex model. We identified a new potential predictor of eye colour and evaluated several widely used advanced machine learning algorithms in predictive analysis of this trait. Our results provide useful hints for developing future predictive models for eye colour in forensic and anthropological studies.

Predicting Eye and Hair Color in a Turkish Population Using the HIrisPlex System

“…Intermediate eye color prediction is difficult, and this can be explained by its complex genetic structure. Some studies have focused on understanding the genetic structure of the intermediate eye color and finding new SNPs [ 23 , 24 ]. Kukla-Bartoszek et al developed an advanced machine learning-based prediction model that increased the sensitivity of the intermediate eye color prediction by up to 39% [ 24 ].…”

“…Some studies have focused on understanding the genetic structure of the intermediate eye color and finding new SNPs [ 23 , 24 ]. Kukla-Bartoszek et al developed an advanced machine learning-based prediction model that increased the sensitivity of the intermediate eye color prediction by up to 39% [ 24 ]. However, current assays and prediction models cannot predict the intermediate eye color with high accuracy.…”

Predicting Physical Appearance from DNA Data—Towards Genomic Solutions

“…Despite its simplicity, the linear model and its generalisations are powerful tools for appearance trait prediction. The theory [ 135 ] and empirical evidence [ 136 , 137 ] suggest that in many cases the dependence between the trait and genetic markers can be captured using linear models. Several studies indicate that they frequently work on par or even better than more complex models, such as ensemble methods or neural networks [ 120 , 132 , 135 , 136 , 137 ], as they are not liable to overfitting.…”

“…The theory [ 135 ] and empirical evidence [ 136 , 137 ] suggest that in many cases the dependence between the trait and genetic markers can be captured using linear models. Several studies indicate that they frequently work on par or even better than more complex models, such as ensemble methods or neural networks [ 120 , 132 , 135 , 136 , 137 ], as they are not liable to overfitting. A distinct advantage of the linear models is their interpretability; the parameter value indicates how the given variable influences the dependent variable for fixed values of the remaining variables.…”

“…Regarding regularisation in the above models, the lasso penalty is the most popular choice, which was successfully used in appearance trait prediction, e.g., in prediction of human height [ 120 ] or eye colour [ 137 ]. The lasso method selects features with non-zero estimated coefficients, and the number selected depends on parameter .…”