AutoSpill: a method for calculating spillover coefficients in high-parameter flow cytometry
Description
Biological utility of AutoSpill. Downstream analyses of data compensated by either the traditional compensation algorithm or AutoSpill. (a) Plots were prepared and compensated using FlowJo v.10.6, using either the default traditional algorithm or uploading the spillover matrix generated by AutoSpill. Representative flow cytometry plots illustrating errors corrected by AutoSpill (first and second column, MM3 dataset; third and fourth column, MM2 dataset). (b-e) All plots were prepared from the same FCS files and compensated using FlowJo v.10.7, using either the traditional algorithm or the AutoSpill option. (b) Hierarchical gating for CD4+CD8+CD25+ lymphocytes, using data compensated by the traditional algorithm or AutoSpill (MM3 dataset). (c) The CD4+CD25+ population was backgated to identify the source of population loss in the traditional algorithm (MM3 dataset). (d) MHCII expression on known negative cells (CD4 T cells), known positive cells (CD11b+ splenocytes), and microglia (MM4 dataset). Percent positive was thresholded using CD4 T cells as the negative. MHCII knockout microglia were used as a “true negative" staining control. (e) Foxp3GFP expression on known bimodal cells (CD4+ splenocytes) and CD11b+ macrophages (MM5 dataset). The positive population was thresholded using the negative CD4 T cell peak. Wildtype mice, without the GFP transgene, were used as a “true negative" staining control. This dataset includes the spillover matrices (traditional and AutoSpill) used in this analysis.