A recent modelling study by Dhole et al. in Evolutionary Applications has thus sought to compare the invasiveness of three gene drives previously described as either spatially or temporally self-limiting.
Recent discussions on gene drives have begun to shift away from highly invasive CRISPR-Cas9 systems toward more controllable alternatives. The study by Dhole et al. focusses on one-locus and two-locus engineered underdominance drives (spatially self-limiting) and the daisy chain drive (temporally self-limiting). Models of each system are formulated and used to compare key behaviours between systems.
First, a baseline performance level is obtained by simulating each system in isolation (i.e. without any migration). This is then compared to a scenario where wild-type individuals migrate into the target population each generation. In doing so, it is found that the performance of each system decreases, with the daisy chain more strongly affected than either underdominance system.
Both underdominance systems possess a threshold frequency above which the system will spread and below which it will be eliminated. With a constant supply of wild-types, underdominance systems require more individuals to be introduced to exceed this threshold, however, once achieved it provides an effective buffer against elimination due to the wild-type influx.
Next the effects of bidirectional migration (i.e. two populations exchange individuals in each generation) are considered. Here it is shown that in most scenarios where the daisy chain drives to high frequency in a target population, it also attains a high frequency in the neighbouring population. For two-locus underdominance systems the spread of the drive is found to be localised to the target population – so long as the migration rate is not too high. This differs from a one-locus underdominance system that could not spread into the neighbouring population under any scenario studied.
Finally, the degree of population suppression given as a by-product of using these systems for population replacement (i.e. where suppression is not the primary goal) is considered. Here it is predicted that, with multiplicative fitness costs, daisy chain drives are likely to produce greater amounts of population suppression than either underdominance-based system.
This is a timely study discussing a major issue associated with potential gene drive releases into the environment. Additionally, the modelling work is kept extremely general and should therefore be useful in considering many different species of interest.