DeMoDa: A global open access database for comparative research in human dental morphological variation

Abstract

Human dental morphology is diverse and varies both within and between populations worldwide. Common variants include different numbers of cusps and roots, as well as different configurations in the fissures, ridges, and grooves on tooth crowns. Because teeth preserve well in taphonomic contexts and retain strong genetic signatures in their morphology, researchers across disciplines use dental form for research ranging from population affinity identification in forensic cases to the reconstruction of population history in archaeological and paleontological studies, and to explore the genetic underpinnings of dental development. However, these analyses are limited, as no publicly available comprehensive database of human dental morphological variation currently exists. Data are typically shared only within professional networks, excluding scientists outside those circles. Information is dispersed across repositories and the supplementary materials that accompany peer-reviewed publications, creating a fragmented and difficult-to-navigate data landscape. Although numberous dental anthropologists devoted their careers to collecting extensive datasets from thousands of individuals worldwide, their data have not yet been published in raw form nor made compatible. Here, we introduce the Dental Morphological Database (DeMoDa), an open access repository comprising 246 dental traits for 17,308 individuals across 32 major geographic regions worldwide spanning the past several thousand years. These data are from the legacy datasets of Turner, Hanihara, Scott, and Irish. Because these researchers used different scoring systems, we developed a scoring system with two steps that enables the three most-common scoring systems to be combined. We provide two versions of the data that correspond to these two steps: one with more nuanced trait scores at the cost of a smaller sample size, and another with less nuanced trait scores but broader sample coverage. We discuss the implications of publishing these legacy data and outline our decision-making process that guided their release in accordance with both the FAIR and CARE principles of open science.