Predictive Modeling of Virus Inactivation by UV

Abstract

UV254 disinfection strategies are commonly applied to inactivate pathogenic viruses in water, food, air, and on surfaces. There is a need for methods that rapidly predict the kinetics of virus inactivation by UV254, particularly for emerging and difficult-to-culture viruses. We conducted a systematic literature review of inactivation rate constants for a wide range of viruses. Using these data and virus characteristics, we developed and evaluated linear and nonlinear models for predicting inactivation rate constants. Multiple linear regressions performed best for predicting the inactivation kinetics of (+) ssRNA and dsDNA viruses, with cross-validated root mean squared relative prediction errors similar to those associated with experimental rate constants. We tested the models by predicting and measuring inactivation rate constants of a (+) ssRNA mouse coronavirus and a dsDNA marine bacteriophage; the predicted rate constants were within 7% and 71% of the experimental rate constants, respectively, indicating that the prediction was more accurate for the (+) ssRNA virus than the dsDNA virus. Finally, we applied our models to predict the UV254 rate constants of several viruses for which high-quality UV254 inactivation data are not available. Our models will be valuable for predicting inactivation kinetics of emerging or difficult-to-culture viruses.