Inferring Accumulation Times of Mitochondrial DNA Deletion Mutants from Cross-Sectional Single-Cell Data: Methodological Framework and Validation

Abstract

The accumulation of mitochondrial DNA (mtDNA) deletion mutants in post-mitotic cells is a hallmark of mammalian ageing and a key contributor to tissue decline in skeletal muscle and neurons. Although the occurrence of such deletions is well documented, the mechanisms by which they clonally expand to levels exceeding the threshold for respiratory chain dysfunction remain unresolved. A transcription-coupled replication model has recently been advanced, predicting that deletions affecting genes involved in a negative feedback mechanism gain a selective replication advantage. This model implies relatively short accumulation times for mutant takeover, a critical but experimentally inaccessible parameter since single-cell measurements are destructive. Here, we present a novel approach to infer such accumulation times from cross-sectional single-cell RNA sequencing (scRNAseq) data, exploiting the fact that mtDNA deletions are also reflected at the transcript level. To establish feasibility, we generated synthetic datasets using two stochastic models of the mitochondrial life cycle and used these as a gold standard. We then applied the Moran process, a classical stochastic model of birth-death dynamics, to calculate distributions of mutant accumulation times and to extract key parameters. The Moran model reproduced the distributions obtained from stochastic simulations with high fidelity, demonstrating robustness across different assumptions about mitochondrial regulation. By fitting the model to synthetic data, we successfully recovered the true values for key parameters like mutation probability, selection advantage, and the fraction of advantageous mutants. Our findings establish a methodological framework for estimating mtDNA mutant dynamics from single-cell transcriptomic data. This approach opens the way for application to large-scale datasets such as Tabula Muris Senis and Tabula Sapiens, offering new insights into the role of mtDNA deletions in ageing and age-related disease.