Extended Data Fig. 2: Supporting information for PS PBMC mtscATAC-seq analyses.

(a) Result of mgatk-del hyperparameter optimization via a simulation framework. The minimum value of the root mean squared error (RMSE) of the estimated and true heteroplasmy is noted with an asterisk over the grid search. (b) Summary of % of cells with 0% heteroplasmy across all hematopoietic cells for the three PS donors. (c) Violin plots of each respective mtDNA deletion for all three patients in selected T cell populations. Black indicates the observed data. Gray represents heteroplasmy under a null model of one mean and the variation attributed to differences in coverage per cell. CD8.TEM and MAIT cells have a bimodal distribution (indicating purifying selection) whereas CD8⁺ naive cells have a distribution that is more consistent with a single mode of heteroplasmy. The percentage of cells with 0% heteroplasmy under observed and null settings are noted for each population below the violins. (d) UMAP visualization of MELAS bridge reference projection across three donors previously reported. (e) Summary of % of cells with 0% heteroplasmy across all hematopoietic cells for the three MELAS donors¹⁰ with refined cell type annotations from the bridge reference projection.