Modeling CRISPR gene drives for suppression of invasive rodents using a supervised machine learning framework.

Area of Science:

• Ecology
• Genetics
• Computational Biology

Background:

• Invasive rodents threaten global biodiversity and native species.
• Traditional control methods are often impractical or costly for island ecosystems.

Purpose of the Study:

• To model the effectiveness of CRISPR gene drive systems for invasive rodent population suppression.
• To utilize machine learning to analyze the complex dynamics and parameter space of gene drive systems.

Main Methods:

• Developed a high-fidelity, spatially explicit, individual-based model of an island rodent population.
• Incorporated three types of suppression gene drive systems with various parameters.
• Created a supervised machine learning meta-model to approximate the outcome space of the population model.

Main Results:

• Gene drive systems have the potential to eliminate island rodent populations under diverse demographic conditions.
• Minimizing resistance allele formation is critical for successful eradication.
• Supervised machine learning efficiently analyzed the complex parameter space and identified key drivers of population dynamics.

Conclusions:

• CRISPR gene drive systems offer a viable strategy for invasive rodent control, particularly on islands.
• Effective management of gene drive systems requires careful consideration of resistance evolution.
• Machine learning is a powerful tool for understanding complex ecological and evolutionary systems.