Answers are grounded in this paper and cite their sources.
Single-cell atlas reveals shared cell states across 18 oral diseases in 365 donors, suggesting pathology arises from distinct recombinations of common cellular programs.
Oral and craniofacial diseases span interconnected epithelial, glandular, stromal, immune, vascular, neural and mineralized-tissue niches, yet single-cell resources remain fragmented by tissue and disease. Here we establish the Human Oral and Craniofacial Atlas (HOCA), integrating 2.2 million quality-controlled cells from 365 donors across anatomical regions and 18 oral disease conditions. By fixing both anatomical position and cell-state identity in one coordinate system, HOCA reframes oral disease not as site-specific pathology or a generic inflammatory response, but as anatomically anchored recombinations of a shared, finite set of cell states. The healthy reference of 0.68 million cells resolves 139 cell types and states, delineating regionalized cellular programmes, recovering rare oral Merkel cells and defining gingiva-enriched basal-interface fibroblast states. Projecting 18 disease conditions onto this reference shows that each is defined by a distinct combination of shared states rather than unique lineages, with a recurrent SPARC+ pericyte fibrovascular node and a CD8 T-cell axis that stratifies disease by transcriptional state rather than abundance. We developed ORACLE (Oral Reference Atlas for Cell Learning and Embedding), a reference-mapping tool transferring HOCA identities to new datasets. The reference and ORACLE are openly available, providing an anatomy-aware coordinate system for oral and craniofacial health and disease.
No takes yet. Share an insight, caveat, or question.
Lin et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: