MutSimABC: A Simulation-Based Approximate Bayesian Computational Framework for Mutation Rate Inference in Long-Lived Trees

Abstract Long-lived trees accumulate somatic mutations over centuries, forming genetic mosaics in which branches carry distinct genotypes shaped by meristem development. Elongation dynamics regulate stem cell lineage maintenance, while branching events redistribute mutations, producing genetic patterns that often diverge from the physical tree topology. Tomimoto and Satake (2023) formalized these processes through mechanistic simulations; however, their framework was designed for forward prediction rather than parameter inference. Estimating mutation rates and developmental parameters from observed data remains computationally intractable for likelihood-based methods. We present MutSimABC, an Approximate Bayesian Computation (ABC) framework that extends the Tomimoto & Satake model to enable simulation-based parameter inference. MutSimABC jointly estimates the mutation rate (μ), elongation parameter (StD), and branching bias (σ) by comparing observed and simulated mutation distributions without requiring explicit likelihood functions. Validation across 169 simulated datasets with known parameters achieved complete recovery of mutation rate and branching bias, and 99.4% recovery for elongation parameters, within 95% highest posterior density (HPD) intervals. Applied to genomic sequencing data from Eucalyptus melliodora, MutSimABC estimated somatic mutation rates ranging from 2.3 × 10−10 to 1 × 10−10 per site per year and inferred partially stochastic meristem dynamics. This framework enables joint inference of mutation and developmental parameters, advancing the mechanistic analysis of somatic evolution in plants, with flexibility that extends to any long-lived organism. Competing Interest Statement The authors have declared no competing interest. Footnotes This work was supported in part by the start-up funds from the University of Auckland, New Zealand under Grant 4020-12090. The code associated with MutSimABC is available at: https://github.com/andreag186/MutSimABC.