Reconstructing human population history from dental phenotypes

Abstract

Dental phenotypic data are often used to reconstruct biological relatedness among past human groups. Teeth are an important data source because they are generally well preserved in the archaeological and fossil record, even when associated skeletal and DNA preservation is poor. Furthermore, tooth form is considered to be highly heritable and selectively neutral; thus, teeth are assumed to be an excellent proxy for neutral genetic data when none are available. However, to our knowledge, no study to date has systematically tested the assumption of genetic neutrality of dental morphological features on a global scale. Therefore, for the first time, this study quantifies the correlation of biological affinities between worldwide modern human populations, derived independently from dental phenotypes and neutral genetic markers. We show that population relationship measures based on dental morphology are significantly correlated with those based on neutral genetic data (on average r = 0.574, p < 0.001). This relatively strong correlation validates tooth form as a proxy for neutral genomic markers. Nonetheless, we suggest caution in reconstructions of population affinities based on dental data alone because only part of the dental morphological variation among populations can be explained in terms of neutral genetic differences.

Variation in dental morphology and inference of continental ancestry in admixed Latin Americans

“…Furthermore, the biodistance analysis and the average estimates of ancestry in the Colombian sample obtained from dental morphological data are consistent with the genetic estimates of ancestry: a predominant European ancestry, with substantial Native American ancestry and a relatively small African contribution. The correlation of certain dental features with specific genetic ancestries suggests that aspects of tooth morphology are likely to be influenced by alleles differentiated in frequency between continental populations (see Hubbard, Guatelli‐Steinberg, & Irish, ; Rathmann et al, ), probably involving loci impacting on tooth development (Dhamo et al, ; Edgar & Ousley, ). The three traits correlated with African ancestry detected here (C5UM2, CNLM1, and C7LM1) have been reported to show markedly higher frequencies in Sub‐Saharan Africans relative to other continental populations (Irish, , ; Scott & Irish, ; Scott & Turner, ).…”

Ancient teeth, phenetic affinities, and African hominins: Another look at where Homo naledi fits in

“…A supplementary technique, such as multidimensional scaling in this study, is then often used to visualize patterning of intersample distances. "It is all but certain that these phenotypic patterns reflect underlying genetic variation" (Rightmire, 1999:2); thus, it is assumed that phenetic similarity provides a reasonable approximation of genetic relatedness, at least based on the findings of recent human studies (e.g., Scott et al, 1983;Larsen, 1997;Scott and Turner, 1997;Martinón-Torres et al, 2007, 2012Hughes and Townsend, 2013a, b;Rathmann et al, 2017).…”

Population history of southern Italy during Greek colonization inferred from dental remains

“…In fact, dental phenotypic data are considered to be most effective at higher geographic scales of study, particularly continental or global (Scott & Turner, ). Recent research has confirmed this and identified congruence in dental phenotypic and neutral genetic datasets from globally distributed populations, with correlations as high as r = 0.635 (Rathmann, Reyes‐Centeno, et al, ). Slightly lower correlations have been found for populations at a regional level, with an agreement of r = 0.500 (Hubbard et al, ).…”

“…More importantly, tooth form has been proposed to be highly heritable and selectively neutral, thus providing an excellent proxy for neutral genetic data (Hillson, ; Scott & Turner, ). In fact, several recent studies have demonstrated the utility of dental phenotypic data in reconstructing genetic relatedness across human populations on different geographic scales (Hubbard, Guatelli‐Steinberg, & Irish, ; Rathmann et al, ) and even between individuals within a population (Paul & Stojanowski, , ; Ricaut et al, ; Stojanowski & Hubbard, ). Finally, dental phenotypic data can be sampled in a nondestructive, cost‐efficient, and straightforward manner using crown width and length measurements (dental metrics) and visual scoring of crown and root shape variants (dental nonmetric traits).…”