WhoseEgg: classification software for invasive carp eggs

Area of Science:

• Ecology
• Invasive Species Management
• Bioinformatics

Background:

• Monitoring invasive carp eggs is crucial for ecological management.
• Genetic identification of fish eggs is accurate but costly and time-consuming.
• Random forest models offer a potentially cheaper alternative using egg morphometrics.

Purpose of the Study:

• To develop a user-friendly application for identifying invasive carp eggs.
• To enable non-R users to utilize random forest models for rapid egg identification.
• To support invasive carp detection in the Upper Mississippi River basin.

Main Methods:

• Development of WhoseEgg, a web-based, point-and-click application.
• Integration of random forest models for egg identification based on morphometric data.
• Facilitation of access to advanced predictive models for resource managers.

Main Results:

• WhoseEgg provides an accessible interface for random forest model application.
• The tool allows rapid and objective identification of invasive carp eggs.
• Demonstrates potential for widespread use in invasive species monitoring.

Conclusions:

• WhoseEgg democratizes the use of advanced predictive models for invasive species management.
• The application facilitates efficient detection of invasive carp (Bighead, Grass, and Silver Carp).
• Future research should focus on expanding the application's capabilities and geographic scope.