Inactivation of B. subtilis spores was calculated as log10 (N/N0),
where N0 was the initial counts in samples and N was the number of
survivors on PEMBA plates after treatments. Spores were assumed
fully inactivated if no colonies were detected on any sample replicates
with the lowest dilution. Tukey's HSD test was used for
comparing means at a 95% significance level. The goodness-of-fit for
each linear regression model was evaluated with the adjusted coefficient
of determination (R2
adj) and root mean squared error (RMSE)
along with P-values derived from the Fisher's F-test. Performance of
nonlinear inactivation models was assessed by root mean square
error (RMSE). Evaluation of R2 as a measure for nonlinear models in
biological studies is not recommended due to overestimated values