Models have made numerous contributions to evolutionary biology, but misunderstandings persist regarding their purpose. By formally testing the logic of verbal hypotheses, proof-of-concept models clarify thinking, uncover hidden assumptions, and spur new directions of study. thumbnail image credit: modified from the Biodiversity Heritage Library

Recent advances in research on gene drives have produced genetic constructs that could theoretically spread a desired gene (payload) into all populations of a species, with a single release in one place. This attribute has advantages, but also comes with risks and ethical concerns. There has been a call for research on gene drive systems that are spatially and/or temporally self‐limiting. Here, we use a population genetics model to compare the expected characteristics of three spatially self‐limiting gene drive systems: one‐locus underdominance, two‐locus underdominance and daisy‐chain drives. We find large differences between these gene drives in the minimum release size required for successfully driving a payload into a population. The daisy‐chain system is the most efficient, requiring the smallest release, followed by the two‐locus underdominance system, and then the one‐locus underdominance system. However, when the target population exchanges migrants with a nontarget population, the gene drives requiring smaller releases suffer from higher risks of unintended spread. For payloads that incur relatively low fitness costs (up to 30%), a simple daisy‐chain drive is practically incapable of remaining localized, even with migration rates as low as 0.5% per generation. The two‐locus underdominance system can achieve localized spread under a broader range of migration rates and of payload fitness costs, while the one‐locus underdominance system largely remains localized. We also find differences in the extent of population alteration and in the permanence of the alteration achieved by the three gene drives. The two‐locus underdominance system does not always spread the payload to fixation, even after successful drive, while the daisy‐chain system can, for a small set of parameter values, achieve a temporally limited spread of the payload. These differences could affect the suitability of each gene drive for specific applications.

Recent advances in research on gene drives have produced genetic constructs that could theoretically spread a desired gene (payload) into all of the populations of a targeted species, with a single release in one place. This attribute has advantages, but also comes with risks and ethical concerns. There has been a call for research on gene drive systems that are spatially and/or temporally self-limiting. Here we use a population genetics model to compare the expected characteristics of three self-limiting gene drive systems: one-locus underdominance, two-locus underdominance, and daisy-chain drive. We find large differences between the selflimiting gene drives in the minimum number of engineered individuals that need to be released for successfully driving a payload into an isolated population. The daisy-chain system is the most efficient, requiring the smallest release, followed by the two-locus underdominance system. The one-locus underdominance system requires the largest releases for successful drive to occur. However, when the target population exchanges migrants with a non-target population, the gene drives requiring smaller releases suffer from higher risks of unintended 2 spread. For payloads that incur relatively low fitness costs (up to 30%), a simple daisy-chain drive is practically incapable of remaining localized, even with migration rates as low as 1% per generation. The two-locus underdominance system can achieve localized spread under a broader range of migration rates and of payload fitness costs, while the one-locus underdominance system largely remains localized. We also find differences in the extent of population alteration and in the permanence of the alteration achieved by the three gene drives. The two-locus underdominance system does not always spread the payload to fixation, even after successful drive, while the daisy-chain system can, in some cases, achieve a temporally-limited spread of the payload. These differences could affect the suitability of each gene drive for specific applications.We ask all readers to recognize that this article has not yet been peer reviewed, and thus, the results shown herein have not yet been validated by researchers other than the authors. We suggest that any reference to or quotation of this article should be made with this recognition.

Optimism regarding potential epidemiological and conservation applications of modern gene drives is tempered by concern about the possibility of unintended spread of engineered organisms beyond the target population. In response, several novel gene drive approaches have been proposed that can, under certain conditions, locally alter characteristics of a population. One challenge for these gene drives is the difficulty of achieving high levels of localized population suppression without very large releases in the face of gene flow. We present a new gene drive system, tethered homing (TH), with improved capacity for both localization and population suppression. The TH drive is based on driving a payload gene using a homing construct that is anchored to a spatially restricted gene drive. We use a proof‐of‐concept mathematical model to show the dynamics of a TH drive that uses engineered underdominance as an anchor. This system is composed of a split homing drive and a two‐locus engineered underdominance drive linked to one part of the split drive (the Cas endonuclease). We use simple population genetic simulations to show that the tethered homing technique can offer improved localized spread of costly transgenic payload genes. Additionally, the TH system offers the ability to gradually adjust the genetic load in a population after the initial alteration, with minimal additional release effort. We discuss potential solutions for improving localization and the feasibility of creating TH drive systems. Further research with models that include additional biological details will be needed to better understand how TH drives would behave in natural populations, but the preliminary results shown here suggest that tethered homing drives can be a useful addition to the repertoire of localized gene drives.

The spread of synthetic gene drives is often discussed in the context of panmictic populations connected by gene flow and described with simple deterministic models. Under such assumptions, an entire species could be altered by releasing a single individual carrying an invasive gene drive, such as a standard homing drive. While this remains a theoretical possibility, gene drive spread in natural populations is more complex and merits a more realistic assessment. The fate of any gene drive released in a population would be inextricably linked to the population's ecology. Given the uncertainty often involved in ecological assessment of natural populations, understanding the sensitivity of gene drive spread to important ecological factors is critical. Here we review how different forms of density dependence, spatial heterogeneity, and mating behaviors can impact the spread of self-sustaining gene drives. We highlight specific aspects of gene drive dynamics and the target populations that need further research. Expected final online publication date for the Annual Review of Ecology, Evolution, and Systematics, Volume 51 is November 2, 2020. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.