Rxivist logo

Saliva cell type DNA methylation reference panel for epidemiology studies in children

By Lauren Y. M. Middleton, John Dou, Jonah Fisher, Jonathan A Heiss, Vy Nguyen, Allan C. Just, Jessica Faul, Erin B. Ware, Colter Mitchell, Justin A Colacino, Kelly M Bakulski

Posted 15 Sep 2020
medRxiv DOI: 10.1101/2020.09.14.20191361

Saliva is a widely used biological sample, especially in pediatric research, containing a heterogenous mixture of immune and epithelial cells. Associations of exposure or disease with saliva DNA methylation can be influenced by cell-type proportions. Here, we developed a saliva cell-type DNA methylation reference panel to estimate interindividual cell-type heterogeneity in whole saliva studies. Saliva was collected from 22 children (7-16 years) and sorted into immune and epithelial cells, using size exclusion filtration and magnetic bead sorting. DNA methylation was measured using the Illumina MethylationEPIC BeadChip. We assessed cell-type differences in DNA methylation profiles and tested for enriched biological pathways. Immune and epithelial cells differed at 164,793 (20.7%) DNA methylation sites (t-test p < 10-8). Immune cell hypomethylated sites mapped to genes enriched for immune pathways (p < 3.2 x 10-5). Epithelial cell hypomethylated sites were enriched for cornification (p = 5.2 x 10-4), a key process for hard palette formation. Saliva immune and epithelial cells have distinct DNA methylation profiles which can drive whole saliva DNA methylation measures. A primary saliva DNA methylation reference panel, easily implemented with an R package, will allow estimates of cell proportions from whole saliva samples and improve epigenetic epidemiology studies by accounting for measurement heterogeneity by cell-type proportions.

Download data

  • Downloaded 482 times
  • Download rankings, all-time:
    • Site-wide: 69,757
    • In epidemiology: 3,086
  • Year to date:
    • Site-wide: 31,641
  • Since beginning of last month:
    • Site-wide: 22,052

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide