Table 2. Regression models of the relationship between somatic instability (SI) in DNA observed in blood and saliva and the estimated progenitor allele length in blood (ePAL) and the age at sampling (Age_s).

Model	Source	Adjusted r2	p	Parameter		Coefficient	Standard error	t-statistic	p
Model 3: log(SI) = β0 + β1 log(ePAL) + β2(Ages) + β3 log(ePAL) * (Ages) + β4 log(ePAL)2 + β5(Ages)2 n = 38 individuals	Blood	0.859	<0.001	Intercept	β0	-28.054	6.599	-4.251	<0.001
				log(ePAL)	β1	21.444	4.404	4.869	<0.001
				Ages	β2	0.179	0.077	2.314	0.027
				log(ePAL) * Ages	β3	-0.049	0.024	-2.012	0.053
				log(ePAL)2	β4	-3.825	0.744	-5.139	<0.001
				Ages2	β5	-0.001	0.000	-2.543	0.016
	Saliva	0.717	<0.001	Intercept	β0	-27.687	8.743	-3.167	0.003
				log(ePAL)	β1	20.471	5.834	3.509	0.001
				Ages	β2	0.224	0.102	2.196	0.035
				log(ePAL) * Ages	β3	-0.068	0.032	-2.120	0.042
				log(ePAL)2	β4	-3.493	0.986	-3.542	0.001
				Ages	β5	-0.001	0.000	-2.080	0.046

The table shows the squared coefficient of correlation (r²) and statistical significance (p) for each tissue, and the coefficient, standard error, t-statistic and statistical significance (p) associated with each parameter in the model. The number of individuals used in the analysis is indicated (n).