A simulation-based deep learning framework for spatially explicit malaria modeling of CRISPR suppression gene drive mosquitoes

Engineered CRISPR gene drives are a promising new strategy for fighting malaria and other vector-borne diseases. One useful approach to predict the outcome of a drive release is individual-based modeling, allowing simulation of the chasing phenomenon. However, the computational demand significantly increases when including many parameters, such as those for disease transmission. To overcome this, we built a deep learning model to understand the effects of different parameters on <i>Anopheles</i> mosquito suppression and human malaria prevalence. Results suggest that reducing embryo resistance, reducing functional resistance, and increasing drive conversion efficiency can contribute to mosquito and malaria suppression. We also observed that the parameter space for eliminating malaria was substantially larger than that for mosquito elimination, suggesting that an imperfect drive may still accomplish its objective despite chasing or functional resistance. Thus, this study shows that suppression gene drives may be highly effective at locally eliminating malaria, even in challenging conditions.