Unifying DNA methylation-based in silico cell-type deconvolution with deconvMe.

Abstract

Summary: Cell-type deconvolution is widely applied to gene expression and DNA methylation data, but access to methods for the latter remains limited. We introduce deconvMe, a new R package that simplifies access to DNA methylation-based deconvolution methods predominantly for blood data, and we additionally compare their estimates to those from gene expression and experimental ground truth data using a unique matched blood dataset.

Availability and implementation: DevonMe is available at https://github.com/omnideconv/deconvMe, the processed blood data is available at https://figshare.com/articles/dataset/methyldeconv_data/28563854/3.