Extended Data Fig. 4: Detection of coding and non-coding genes in PBMCs.
From: Scalable single-cell total RNA sequencing unifies coding and noncoding transcriptomics
a. Pseudobulk comparison of protein-coding gene expression between TotalX and 10x-dUTSO across PBMC cell types. Scatterplots show pseudobulk expression levels of protein-coding genes for each recovered PBMC cell type. Genes are colored by highly variable status, and transcripts with the largest differences between methods are labeled. b. Number of genes detected per RNA biotype (minimum expression in ≥3 cells) in the TotalX PBMC dataset. Biotypes include protein-coding genes, lncRNAs, tRNAs, snoRNAs, snRNAs, histone RNAs, miRNAs, and miscellaneous RNAs, reflecting TotalX’s capacity to capture broad transcript classes. c. Cell-type specific differentially expressed genes (DEGs) stratified by RNA biotype. Bars indicate the number of DEGs (FDR < 0.01) for each cell type and RNA class. While protein-coding genes remain dominant, multiple non-coding biotypes exhibit strong cell-type–specific expression – for example, lncRNAs in monocytes, miRNAs in lymphocytes, and snRNAs in plasmacytoid dendritic cells (pDCs). d. Relationship between amino acid demand (based on codon usage of expressed genes) and tRNA supply (measured by TotalX) across immune cell types. The strong correlation (Pearson r = 0.79, p = 1.59e–66) confirms internal consistency and biological relevance of captured tRNA profiles. Selected amino acids such as Arginine and Glycine show relative oversupply, whereas Tryptophan and Phenylalanine are undersupplied.
