Kinetic data analysis for probiotic Lacticaseibacillus rhamnosus GG growth and pH drop in rice-based milk alternative

The objective of this study was to describe and predict the behavior of probiotic Lacticaseibacillus rhamnosus GG (LGG) in a rice-based milk alternative over a temperature range of 12 to 44 °C using the standard three-step kinetic data analysis process. According to the root-mean-square error (RMSE), the primary Baranyi and Roberts (BR) and the new logistic (NL) models described population growth with nearly identical fitting accuracy, with averages of 0.112 and 0.115, respectively. The secondary model for the specific growth rate (μ_max), based on the outputs of the primary models (BR and NL), also produced very similar estimates of T _min values at 4.61 °C and 4.71 °C and T _max values at 49.67 °C and 49.69 °C, respectively. In general, the average coefficient of determination was R ² = 0.991 and RMSE = 0.185 for predicting LGG growth using the BR and NL models, each incorporating secondary models for the lag phase and μ _max (tertiary models). To model the pH kinetics, which exhibited a sigmoidal curve pattern, an approach analogous to that used for growth prediction was applied. The final pH prediction was performed using the Geeraerd model (GM_{pH, T}), integrated with both secondary models and cardinal growth temperatures of T _min = 4.61 °C and T _max = 49.67 °C. This tertiary model demonstrated strong statistical performance, with R ² = 0.941 and RMSE = 0.386. The ability to predict the behavior of L. rhamnosus GG is valuable for designing and optimizing dairy-free matrices with probiotic potential.