Mixed-effects logistic regression model with interactions.

Fixed effects: age category (mother, pup), tissue (brain, muscle, oocyte pool, single oocyte [combined across all oocytes measured for a mouse]), mtDNA compartment (D-loop, non–D-loop), mutation type (transversion, A>G and T>C transitions, C>T and G>A transitions), and interactions between mutation type (transversion, A>G and T>C transitions, C>T and G>A transitions) and each of the other variables. Baseline for fixed effects: age category, mother; tissue, brain; mtDNA compartment, D-loop; mutations type, transversion. Random effects: mouse ID (multiple observations are related to the same individual, 36 individuals in total). Response: mutation frequency, with weights given by the number of nucleotides corresponding to each observation. Marginal pseudo-R2 (represents the variance explained by the fixed effects): 23.31%. Conditional pseudo-R2 (represents the variance explained by both fixed and random effects): 24.02%. Interactions are significant (chi-squared test for comparison between full and reduced model): p-value = 1.3e−30. ID, identifier; mtDNA, mitochondrial DNA.
(XLSX)