Decay of individual strains of bacterial viruses in laboratory settings is generally exponential (De Paepe & Taddei, 2006). In situ measurements of bacteriophage survivorship, however, have suggested that phage communities exhibit long-tailed non-exponential decay (Hoffman et al., 2007). We hypothesize that such long-tailed non-exponential decay rates results from sampling mixed communities of phages with varying decay rates, and can thus be deconstructed into a set of exponentially-decaying phage strains. Our objective is thus to create a theoretical method to deconstruct community decay rates into their exponential components and to test the efficacy of this method on communities of mixed phages with known decay rates.
Results/Conclusions
Our theoretical method inverts a community decay curve, which estimates the mortality rate and the population size of each component strain. This method is exact in the absence of noise, and we have developed additional mathematical techniques, based on the principles of regularization, to estimate component exponential rates given noisy data. In order to test our hypothesis regarding community decay, we isolated two-strain communities of phages with known decay rates at fixed initial proportions. Preliminary results from community decay experiments suggests that we can invert these community decay curves and compare the estimated mortality rates of individual strains to mortality rates when measured in isolation. We discuss the potential application of our method to inferring the functional traits of unculturable environmental phages.
