Extended Data Fig. 3: Comparative profiling of PBMC subtypes across protocols and sensitivity of monocytes to enzymatic polyadenylation.

a. Dot plot showing expression of canonical marker genes across annotated immune cell types, including major T cell subsets, monocytes, NK cells, dendritic cells, basophils, plasmablasts, and platelets. Marker genes were used for manual annotation of cell clusters in TotalX, 10x-dUTSO, and TotalX+miRNA(+) protocols. b. UMAP of PBMCs colored by protocol, demonstrating that TotalX and TotalX + retain consistent clustering and recover all expected immune populations observed in the 10x-dUTSO control. c. Left: Bubble plot of relative frequencies of immune cell types across protocols. Right: Absolute numbers of cells recovered by TotalX, TotalX pA + + and 10x-dUTSO. TotalX recovers comparable or greater numbers of low-RNA-content populations (for example, platelets, basophils, proerythroblasts). d. Violin plots showing the number of genes (left) and UMIs (right) detected per cell type. Low-content cell types such as platelets and basophils show improved gene detection under TotalX, while other major cell types are comparable across protocols. e. UMAP focused on monocyte populations reveals a distinct subcluster of CD14+ monocytes in TotalX + showing transcriptional divergence, indicative of sensitivity to protocol conditions. f. Dot plot showing selective upregulation of JARID2, GAB2, and APOBEC3A, along with downregulation of canonical monocyte markers (for example, LYZ, CD14), in CD14+ monocytes exposed to increasing concentrations of ATP and polyA polymerase. These cells also show enrichment for apoptosis and chromatin remodeling genes. g. Violin plots showing global decrease in gene detection and increased mitochondrial RNA content in the sensitive monocyte subpopulation, consistent with stress or partial identity loss under certain enzymatic conditions.