Gene drives and population persistence vs elimination: The impact of spatial structure and inbreeding at low density

Genetic and geographic population structure in the malaria vector, Anopheles farauti, provides a candidate system for pioneering confinable gene-drive releases

Indoor insecticide applications are the primary tool for reducing malaria transmission in the Solomon Archipelago, a region where Anopheles farauti is the only common malaria vector. Due to the evolution of behavioural resistance in some An. farauti populations, these applications have become less effective. New malaria control interventions are therefore needed in this region, and gene-drives provide a promising new technology. In considering developing a population-specific (local) gene-drive in An. farauti, we detail the species' population genetic structure using microsatellites and whole mitogenomes, finding many spatially confined populations both within and between landmasses. This strong population structure suggests that An. farauti would be a useful system for developing a population-specific, confinable gene-drive for field release, where private alleles can be used as Cas9 targets. Previous work on Anopheles gambiae has used the Cardinal gene for the development of a global population replacement gene-drive. We therefore also analyse the Cardinal gene to assess whether it may be a suitable target to engineer a gene-drive for the modification of local An. farauti populations. Despite the extensive population structure observed in An. farauti for microsatellites, only one remote island population from Vanuatu contained fixed and private alleles at the Cardinal locus. Nonetheless, this study provides an initial framework for further population genomic investigations to discover high-frequency private allele targets in localized An. farauti populations. This would enable the development of gene-drive strains for modifying localised populations with minimal chance of escape and may provide a low-risk route to field trial evaluations.

Incorporating ecology into gene drive modelling

Gene drive technology, in which fast-spreading engineered drive alleles are introduced into wild populations, represents a promising new tool in the fight against vector-borne diseases, agricultural pests and invasive species. Due to the risks involved, gene drives have so far only been tested in laboratory settings while their population-level behaviour is mainly studied using mathematical and computational models. The spread of a gene drive is a rapid evolutionary process that occurs over timescales similar to many ecological processes. This can potentially generate strong eco-evolutionary feedback that could profoundly affect the dynamics and outcome of a gene drive release. We, therefore, argue for the importance of incorporating ecological features into gene drive models. We describe the key ecological features that could affect gene drive behaviour, such as population structure, life-history, environmental variation and mode of selection. We review previous gene drive modelling efforts and identify areas where further research is needed. As gene drive technology approaches the level of field experimentation, it is crucial to evaluate gene drive dynamics, potential outcomes, and risks realistically by including ecological processes.

Dispersal inference from population genetic variation using a convolutional neural network

The geographic nature of biological dispersal shapes patterns of genetic variation over landscapes, making it possible to infer properties of dispersal from genetic variation data. Here, we present an inference tool that uses geographically distributed genotype data in combination with a convolutional neural network to estimate a critical population parameter: the mean per-generation dispersal distance. Using extensive simulation, we show that our deep learning approach is competitive with or outperforms state-of-the-art methods, particularly at small sample sizes. In addition, we evaluate varying nuisance parameters during training-including population density, demographic history, habitat size, and sampling area-and show that this strategy is effective for estimating dispersal distance when other model parameters are unknown. Whereas competing methods depend on information about local population density or accurate inference of identity-by-descent tracts, our method uses only single-nucleotide-polymorphism data and the spatial scale of sampling as input. Strikingly, and unlike other methods, our method does not use the geographic coordinates of the genotyped individuals. These features make our method, which we call "disperseNN," a potentially valuable new tool for estimating dispersal distance in nonmodel systems with whole genome data or reduced representation data. We apply disperseNN to 12 different species with publicly available data, yielding reasonable estimates for most species. Importantly, our method estimated consistently larger dispersal distances than mark-recapture calculations in the same species, which may be due to the limited geographic sampling area covered by some mark-recapture studies. Thus genetic tools like ours complement direct methods for improving our understanding of dispersal.

Leveraging eco-evolutionary models for gene drive risk assessment

Engineered gene drives create potential for both widespread benefits and irreversible harms to ecosystems. CRISPR-based systems of allelic conversion have rapidly accelerated gene drive research across diverse taxa, putting field trials and their necessary risk assessments on the horizon. Dynamic process-based models provide flexible quantitative platforms to predict gene drive outcomes in the context of system-specific ecological and evolutionary features. Here, we synthesize gene drive dynamic modeling studies to highlight research trends, knowledge gaps, and emergent principles, organized around their genetic, demographic, spatial, environmental, and implementation features. We identify the phenomena that most significantly influence model predictions, discuss limitations of biological complexity and uncertainty, and provide insights to promote responsible development and model-assisted risk assessment of gene drives.

The Promise and Challenge of Genetic Biocontrol Approaches for Malaria Elimination

Malaria remains an ongoing public health challenge, with over 600,000 deaths in 2021, of which approximately 96% occurred in Africa. Despite concerted efforts, the goal of global malaria elimination has stalled in recent years. This has resulted in widespread calls for new control methods. Genetic biocontrol approaches, including those focused on gene-drive-modified mosquitoes (GDMMs), aim to prevent malaria transmission by either reducing the population size of malaria-transmitting mosquitoes or making the mosquitoes less competent to transmit the malaria parasite. The development of both strategies has advanced considerably in recent years, with successful field trials of several biocontrol methods employing live mosquito products and demonstration of the efficacy of GDMMs in insectary-based studies. Live mosquito biocontrol products aim to achieve area-wide control with characteristics that differ substantially from current insecticide-based vector control methods, resulting in some different considerations for approval and implementation. The successful field application of current biocontrol technologies against other pests provides evidence for the promise of these approaches and insights into the development pathway for new malaria control agents. The status of technical development as well as current thinking on the implementation requirements for genetic biocontrol approaches are reviewed, and remaining challenges for public health application in malaria prevention are discussed.

Gene drive designs for efficient and localisable population suppression using Y-linked editors

The sterile insect technique (SIT) has been successful in controlling some pest species but is not practicable for many others due to the large number of individuals that need to be reared and released. Previous computer modelling has demonstrated that the release of males carrying a Y-linked editor that kills or sterilises female descendants could be orders of magnitude more efficient than SIT while still remaining spatially restricted, particularly if combined with an autosomal sex distorter. In principle, further gains in efficiency could be achieved by using a self-propagating double drive design, in which each of the two components (the Y-linked editor and the sex ratio distorter) boosted the transmission of the other. To better understand the expected dynamics and impact of releasing constructs of this new design we have analysed a deterministic population genetic and population dynamic model. Our modelling demonstrates that this design can suppress a population from very low release rates, with no invasion threshold. Importantly, the design can work even if homing rates are low and sex chromosomes are silenced at meiosis, potentially expanding the range of species amenable to such control. Moreover, the predicted dynamics and impacts can be exquisitely sensitive to relatively small (e.g., 25%) changes in allele frequencies in the target population, which could be exploited for sequence-based population targeting. Analysis of published Anopheles gambiae genome sequences indicates that even for weakly differentiated populations with an FST of 0.02 there may be thousands of suitably differentiated genomic sites that could be used to restrict the spread and impact of a release. Our proposed design, which extends an already promising development pathway based on Y-linked editors, is therefore a potentially useful addition to the menu of options for genetic biocontrol.

A theory of resistance to multiplexed gene drive demonstrates the significant role of weakly deleterious natural genetic variation

Evolution of resistance is a major barrier to successful deployment of gene-drive systems to suppress natural populations, which could greatly reduce the burden of many vector-borne diseases. Multiplexed guide RNAs (gRNAs) that require resistance mutations in all target cut sites are a promising antiresistance strategy since, in principle, resistance would only arise in unrealistically large populations. Using stochastic simulations that accurately model evolution at very large population sizes, we explore the probability of resistance due to three important mechanisms: 1) nonhomologous end-joining mutations, 2) single-nucleotide mutants arising de novo, or 3) single-nucleotide polymorphisms preexisting as standing variation. Our results explore the relative importance of these mechanisms and highlight a complexity of the mutation-selection-drift balance between haplotypes with complete resistance and those with an incomplete number of resistant alleles. We find that this leads to a phenomenon where weakly deleterious naturally occurring variants greatly amplify the probability of multisite resistance compared to de novo mutation. This key result provides design criterion for antiresistance multiplexed systems, which, in general, will need a larger number of gRNAs compared to de novo expectations. This theory may have wider application to the evolution of resistance or evolutionary rescue when multiple changes are required before selection can act.