A novel cell-type deconvolution algorithm reveals substantial contamination by immune cells in saliva, buccal and cervix.

Aim An outstanding challenge in epigenome studies is the estimation of cell-type proportions in complex epithelial tissues. Materials & methods Here, we construct and validate a DNA methylation reference and algorithm for complex tissues that contain epithelial, immune and nonimmune stromal cells. Results Using this reference, we show that easily accessible tissues such as saliva, buccal and cervix exhibit substantial variation in immune cell (IC) contamination. We further validate our reference in the context of oral cancer, where it correctly predicts an increased IC infiltration in cancer but suppressed in patients with highest smoking exposure. Finally, our method can improve the specificity of differentially methylated CpG calls in epithelial cancer. Conclusion The degree and variation of IC contamination in complex epithelial tissues is substantial. We provide a valuable resource and tool for assessing the epithelial purity and IC contamination of samples and for identifying differential methylation in such complex tissues.