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

Area of Science:

• Genetics
• Vector Biology
• Computational Biology

Background:

• CRISPR gene drives offer a novel strategy for controlling vector-borne diseases like malaria.
• Individual-based modeling is useful for predicting gene drive outcomes but can be computationally intensive.
• Integrating numerous parameters, including disease transmission dynamics, further complicates these models.

Purpose of the Study:

• To develop a deep learning model for analyzing the impact of various parameters on Anopheles mosquito suppression.
• To assess the influence of these parameters on human malaria prevalence.
• To identify key factors driving the success of gene drives in malaria elimination.

Main Methods:

• Utilized a deep learning approach to model complex gene drive dynamics.
• Simulated Anopheles mosquito population suppression based on varying genetic and environmental parameters.
• Analyzed the relationship between mosquito suppression and human malaria prevalence.

Main Results:

• Reducing embryo and functional resistance, alongside increasing drive conversion efficiency, significantly aids mosquito and malaria suppression.
• The parameter space for malaria elimination was found to be considerably larger than for mosquito elimination.
• Imperfect gene drives can still achieve malaria elimination objectives despite resistance mechanisms.

Conclusions:

• Suppression gene drives demonstrate high potential for local malaria elimination, even in difficult scenarios.
• Deep learning models provide an efficient method to explore complex gene drive parameter spaces.
• Optimizing drive efficiency and overcoming resistance are critical for successful malaria vector control.