A discrete-time epidemiological model to quantify selection for disease resistance

A genetic epidemiological model (GEM) for investigating the effect of selec tion for disease resistance on the Epidemiology of infectious diseases is p resented and applied to a pig breeding scenario. Fundamental to the model i s R-0, the basic reproductive ratio. R-0 is the expected number of secondar y infections caused by a single infection. If R-0 is greater than 1, there will be an epidemic. The nim of the model is to quantify the effect of sele ction on R-0 and the consequences this has on disease epidemiology. Two imp lementations are presented: selection for reduced susceptibility/infectivit y to a disease and introgression of a major resistance gene. The results su ggest that the effects of selection for reduced susceptibility/infectivity are critically dependent on the infectiousness of the disease. Under the as sumptions made in the model, for a disease with a low infection level, it t akes approximately 25 years of selection until R-0 is less than 1 and the p opulation is safe from epidemics should the infectious agent be present. Fo r a highly infectious disease, this rime may be as long as 100 years. For g ene introgression, the population is expected to be free from epidemics wit hin 5 years and the time to reduce R-0 to less than 1 is largely independen t of the disease being considered. With gene introgression, the proportion of the population which needs to be resistant to ensure that R-0 is less th an one is shown to be a function of the initial R-0 for the disease. Althou gh selection, as modelled, results in a linear decline in R-0, the reductio n in the proportion of animals infected during an epidemic is non-linear. T he selection process reduces the amount of infectious material that is in t he environment when an infection occurs and this decreases the force of inf ection on unselected animals. This phenomenon results in a marked interacti on between host genotype and disease epidemiology. Thus, the results of the model show that altering the genetics of individual animals affects the ep idemiology of the disease at the population level. The model can be applied to any farm structure and any microparasitic infections disease.