Abstract
One of the pieces of information required for the biological profile is ancestry, providing an estimation of the geographical origin of an unknown individual. In the last years, online software techniques have been developed for practitioners to use as tools for identification. The present project aims to evaluate the applicability of one traditional method and two online software applications for ancestry estimation through skull and mandible metrics on a Spanish modern sample.
The sample consists of 114 adult individuals from the School of Legal Medicine in Madrid (Del Rio-Muñoz, 2000). Thirty skull measurements and nine mandibular measurements were inserted into Giles and Elliot (1962) traditional formulae, and into two online software tools, Ancestrees and (hu)MANid. Methods’ reliability consisted of the calculation of classification rates accounting for total and sex-specific accuracies, as well as software statistical output.
Giles and Elliot’s (1962) method performed similarly for males and females, with overall accuracy of 90%. AncesTrees performed with the greatest accuracy classifying 86.05% as European and 54.56% as Southwestern European when different algorithms are applied. (hu)MANid correctly classified 70.59-80.00%, when correct sex assessment was not taken into account. Accounting for sex and ancestry estimation, accuracy rates ranged from 62.75-80.00%.
Validation studies on traditional and new methods are required to substantiate results. Our findings demonstrate that reference populations, admixture, and intrinsic and extrinsic factors can potentially influence the methods accuracy and reliability. Future research should focus on the inclusion of underrepresented groups, especially in view of the current humanitarian migratory crisis.
The sample consists of 114 adult individuals from the School of Legal Medicine in Madrid (Del Rio-Muñoz, 2000). Thirty skull measurements and nine mandibular measurements were inserted into Giles and Elliot (1962) traditional formulae, and into two online software tools, Ancestrees and (hu)MANid. Methods’ reliability consisted of the calculation of classification rates accounting for total and sex-specific accuracies, as well as software statistical output.
Giles and Elliot’s (1962) method performed similarly for males and females, with overall accuracy of 90%. AncesTrees performed with the greatest accuracy classifying 86.05% as European and 54.56% as Southwestern European when different algorithms are applied. (hu)MANid correctly classified 70.59-80.00%, when correct sex assessment was not taken into account. Accounting for sex and ancestry estimation, accuracy rates ranged from 62.75-80.00%.
Validation studies on traditional and new methods are required to substantiate results. Our findings demonstrate that reference populations, admixture, and intrinsic and extrinsic factors can potentially influence the methods accuracy and reliability. Future research should focus on the inclusion of underrepresented groups, especially in view of the current humanitarian migratory crisis.
| Original language | English |
|---|---|
| Publication status | Published - 17 Oct 2021 |
| Event | 2nd International and 18th National Forensic Sciences Congress (Turkey) - Duration: 14 Oct 2021 → 17 Oct 2021 https://adlibilimler2021.com/en/3066-scientific-program-online |
Conference
| Conference | 2nd International and 18th National Forensic Sciences Congress (Turkey) |
|---|---|
| Period | 14/10/21 → 17/10/21 |
| Internet address |