The 89th Annual Meeting of the American Association of Physical Anthropologists (2020)


How good are ASUDAS traits for assessing population relatedness? An answer—from comparisons of African dental and genetic data

JOEL D. IRISH, ADELINE MOREZ and LINUS GIRDLAND FLINK.

Anthropology and Archaeology, School of Biological and Environmental Sciences, Liverpool John Moores University

April 17, 2020 , Platinum Ballroom Add to calendar

Arizona State University Dental Anthropology System (ASUDAS) nonmetric traits are often considered the “gold standard” for estimating phenetic affinities at individual through global scales. Beyond user-friendliness, high observer replicability, low cost, and non-destructive sampling, these traits are largely selectively neutral with a high genetic component in expression. Thus, they are deemed expedient proxies for genetic markers. But are they really?

In answer, several recent studies compared ASUDAS and genetic (SNPs, STRs) data via model-free and model-bound methods at local, regional, and global levels among recent peoples. The consensus is that the nonmetric data yield dependable results, although some authors are less convinced than others.

We assess further the utility of ASUDAS traits—here on a continental scale. Using a model-free approach for matched samples of 12 North and sub-Saharan African populations, mean measure of divergence (MMD) distances from 36 traits were compared with those of Hudson Fst based on high density SNP (>300,000) data. These matrices are strongly correlated (rm=0.786, p=9.999e-05). After trait editing, described elsewhere by the first author, a 25-trait MMD and Fst correlation increased (rm=0.838, p=9.999e-05). This concordance is notably better than in the abovementioned studies. Of further interest, the correlation for both 36- and 25-trait MMD with geographic distances is higher (rm=0.700, p=9.999e-05) than for Fst and geography (rm=0.486, p=0.001).

These findings suggest ASUDAS traits are valuable for assessing relatedness, and several explanations for the improved concordance are presented. This and our current global level research continues to address just how good these traits are.

Data collection by the first author was funded by the National Science Foundation (BNS-9013942, BNS-0104731, BCS-0840674), the ASU Research Development Program, and the American Museum of Natural History.


Slides/Poster (pdf)