Bridging machine learning and evolutionary optimization of threshold specific dosages of Nisin to suppress MRSA biofilm.

Methicillin resistant Staphylococcus aureus (MRSA), a Gram-positive potent biofilm forming pathogen responsible for minor skin infection to life-threatening sepsis due to its resistance towards traditional antibiotics. The biofilm forming ability of this organism serves as a primary driver of this resistance, rendering traditional therapeutic strategies critically limited. To address this challenge, a natural antimicrobial peptide, Nisin, produced by Lactococcus lactis, was deployed employed against 14 different MRSA isolates. The present study implements an artificial intelligence and machine learning (AI-ML) based predictive framework for optimizing the dosing regimens of Nisin for maximized biofilm inhibition under tailored conditions. Furthermore, to map the treatment dynamics, an empirical dataset of 204 in vitro observations was generated across three moving parameters (Concentration of Nisin, Initial inoculum density adjusted to CFU/mL, and Incubation time). Six different predictive regressor models including multiple linear regression (MLR), polynomial regression (PR), support vector regression (SVR), response surface methodology (RSM), artificial neural network (ANN) configured as an artificial neural network regressor (ANNR) were thoroughly evaluated. Amongst them, the 4th degree PR model demonstrated the superior predictive performance with a R2 value of 0.964 on testing vectors, and was subsequently incorporated with Genetic Algorithm (GA) to achieve an optimal therapeutic matrix [ Concentration of Nisin = 40 µg/mL, Initial inoculum density adjusted to 1.2 × 105 CFU/mL, and Incubation period 7.63 h]. This targeted window effectively maps a clinically relevant, early-stage MRSA infection scenario. Additionally, these optimized parameters were further validated through several antibiofilm assay experimental observations. Collectively, this study establishes an AI driven, precision-guided approach to combat MRSA-borne infection.