Trait-based prediction of disease-vector mosquito invasion potential

Area of Science:

• Ecology
• Epidemiology
• Biogeography

Background:

• Mosquito-borne diseases are a growing global health concern.
• Human activities facilitate the introduction of invasive mosquito vectors.
• Established invasive mosquitoes are difficult to control, necessitating early detection.

Purpose of the Study:

• To develop a predictive framework for mosquito invasion potential.
• To identify key ecological and life-history traits associated with invasion.
• To prioritize species for proactive surveillance and control efforts.

Main Methods:

• Compiled a dataset of 26 traits for 184 medically important mosquito species.
• Utilized random forest models to predict introduction and establishment probabilities.
• Analyzed traits related to geographic origin, breeding sites, and environmental tolerance.

Main Results:

• Models demonstrated moderate to good predictive performance (AUC = 0.78-0.85).
• Species from Asia and Australia, adapted to human-made sites, and tolerant of extreme environments showed higher invasion risk.
• Identified 24 species with high predicted introduction probability, 17 of which also had high establishment probability.

Conclusions:

• Mosquito invasion potential can be accurately predicted using species traits.
• This framework provides a quantitative basis for proactive surveillance.
• Prioritization of high-risk species can mitigate future introductions and disease transmission.