DA between groups of cells was computed by constructing a logistic regression model predicting group membership based on the accessibility of a given peak in the set of cells being compared, with the total number of counts in each cell included as a latent variable in the model and comparing this with a null model using a likelihood ratio test. This was performed using the FindMarkers function in Seurat v3.2.0, with test.use = ‘LR’ and latent.vars = ‘nCount_ATAC’. We classified peaks with an adjusted P value (Bonferroni-corrected) <0.01 and absolute log2 fold change >0.4 as being DA between the cell groups.
