Mathematical modeling of genetic pest management through female lethality with independently segregating alleles

Many novel genetic approaches are under development to combat insect pests, which continue to impose substantial burdens in the forms of transmission of vector-borne disease and the reduction of agricultural yields. One category of genetic pest management aims to suppress or locally eliminate a species through large, repeated releases of fertile males that carry a dominant lethal genetic construct to render all, or exclusively female, offspring unviable. In all strains that have been tested in the field, the genetic components are in a single construct inserted at a unique location in the genome. Other conditional female lethal strains have been developed with two components at independently segregating loci. It has been suggested that males from these strains only be used for a sterile release program as the components are not linked. Here, we use a population genetics model with density-dependence to evaluate the relative effectiveness of female killing approaches based on single- and multi-locus arrangements of the components. We find that, in general, the single-locus arrangement results in slightly faster population suppression, but the two-locus arrangement can eventually cause stronger suppression and cause local extinction with a smaller release size. Based on our results, there is no a priori reason that males carrying two or more independently segregating components need to be sterilized prior to release. In some cases, a fertile release would be more efficient for population suppression.