Table 3.
Average identification and verification results for the EURO cohort
| EURO Cohort | EER | σ | AUC | σ | R1 | σ | R10 | σ | R20 | σ |
|---|---|---|---|---|---|---|---|---|---|---|
| SEX | 34.7 | 1.2 | 73.5 | 1.3 | 1.35 | 0.00 | 19.64 | 0.04 | 39.96 | 0.08 |
| SNPS | 40.2 | 2.1 | 63.6 | 1 | 2.48 | 0.98 | 21.90 | 1.08 | 38.49 | 0.59 |
| GB | 42.7 | 1.8 | 60.7 | 0.5 | 1.35 | 0.00 | 17.05 | 2.53 | 30.93 | 1.42 |
| BMI | 38.5 | 1.9 | 66.1 | 1.1 | 1.24 | 0.19 | 18.17 | 0.41 | 34.88 | 1.24 |
| AGE | 38.9 | 1.8 | 67.1 | 1 | 1.35 | 0.00 | 19.53 | 0.18 | 37.92 | 0.74 |
| SEX + SNPS | 29.9 | 1.8 | 79 | 1.6 | 3.72 | 1.21 | 34.09 | 1.87 | 56.21 | 2.55 |
| SEX + SNPS + GB | 28.6 | 1.9 | 80.8 | 1.6 | 5.19 | 0.70 | 38.60 | 3.02 | 58.92 | 3.21 |
| SEX + SNPS + GB + BMI | 25.2 | 1.8 | 84 | 1.5 | 5.30 | 0.51 | 45.26 | 3.51 | 69.87 | 3.78 |
| SEX + SNPS + GB + BMI + AGE | 20.6 | 2.1 | 88 | 1.4 | 8.92 | 0.97 | 56.66 | 3.45 | 81.27 | 1.67 |
| SEX + GB + BMI + AGE | 22.1 | 1.7 | 86.3 | 1.3 | 5.31 | 0.01 | 50.23 | 3.72 | 75.51 | 1.55 |
| X = [SEX, AGE, BMI, GB] | 25.6 | 0.8 | 81.8 | 0.6 | 5.64 | 1.41 | 44.36 | 2.34 | 68.62 | 2.54 |
| X = [SEX, AGE, BMI, GB, SNPS] | 25.4 | 0.8 | 82.5 | 0.5 | 6.32 | 0.52 | 48.20 | 0.80 | 70.88 | 0.29 |
The results for a DNA-to-Face prediction strategy from a set of predictors X are given in the last two rows. All values are given in percentage. For the EURO cohort, the performances of the DNA-to-Face prediction and Face-to-DNA classification setups are shown with and without the contribution of the SNPs from Table 1. Random performance is given as EER = 0.5, AUC = 0.5, R1 = 1%, R10 = 10%, R20 = 20%. % refers to the percentage of individuals in the gallery (n = 295, test dataset)
EER verification equal error rate, AUC verification area under the curve, R1 rank 1% identification rate, R10 rank 10% identification rate, R20 rank 20% identification rate, σ standard deviation, GB genomic background, BMI body mas index