Deconvolution was performed with the ReferenceBasedDecomposition function of the R package BisqueRNA (v.1.0.4)70, using the use.overlap = FALSE option. The single-cell reference dataset used was the single-nucleus RNA-seq from the 10X protocol, which includes tissue from eight donors and five brain regions26. The ten cell types considered in the deconvolution of the tissue were astrocytes, endothelial cells, microglia, macrophages, mural cells, oligodendrocytes, OPCs, T cells, excitatory neurons and inhibitory neurons. Marker genes were selected by first filtering for genes common between the bulk and reference data and then calculating the ratio of the mean expression of each gene in the target cell type over the highest mean expression of that gene in a nontarget cell type. The 25 genes with the highest ratios for each cell type were selected as markers.
