Longitudinal somatic mutation analysis of individual human CD8⁺ T cells by a UMIbased single cell mutation detection (USCMD) method

Somatic mutation is a contributing factor for cancer and aging. DNA-based single-cell mutation detection methods have been developed, but are hindered by the limited number of cells examined as well as by errors from PCR amplification and sequencing methods. Here, we developed a UMI-based scRNAseq mutation detection (USCMD) method that allows for the examination of up to 10,000 cells per sample with reduced PCR and sequencing errors, and which also filters potential errors of RNA polymerase II. We applied USCMD to analyze the scRNAseq data from human CD8 +T cells collected from 8 donors with 2 visits over a period of 9 years (aged from 30 to 69 years at first visit). We found that some novel missense mutations were associated with changes (increase or decrease) in the number of cells carrying them over time, and were able to identify clonally related cells based on shared somatic mutations. Together, this USCMD method improves the accuracy of mutational analysis of scRNAseq data and allows parallel transcriptional assessment of mutated and unmutated cells to understand their potential functional consequences.