QuantiFly: Robust Trainable Software for Automated Drosophila Egg Counting.

Area of Science:

• * Genetics and Genomics
• * Bioinformatics and Computational Biology
• * Animal Behavior and Physiology

Background:

• * Accurate quantification of Drosophila egg laying is crucial for assessing fitness in many laboratory settings.
• * Current manual counting methods are time-consuming, tedious, and prone to human error, especially for large-scale experiments.
• * There is a need for automated, efficient, and accurate tools to streamline this process.

Purpose of the Study:

• * To develop and validate QuantiFly, an automated software tool for quantifying Drosophila egg laying.
• * To improve upon existing pattern recognition and machine-learning algorithms for egg counting.
• * To provide an accessible, user-friendly, and open-source solution for researchers.

Main Methods:

• * Development of the QuantiFly software, incorporating a refined pattern recognition and machine-learning algorithm with bias correction.
• * Implementation of an intuitive graphical user interface for ease of use.
• * Testing QuantiFly against manual counts using digital images of eggs on transparent and opaque media.

Main Results:

• * QuantiFly achieved average accuracies of 94% on transparent media and 85% on opaque media.
• * The software demonstrated robustness against food surface artifacts like bubbles and crevices.
• * QuantiFly significantly reduces counting time, offering advantages for experiments with as few as 20 vials.

Conclusions:

• * QuantiFly is a highly accurate and efficient automated tool for quantifying Drosophila egg laying.
• * The software provides a substantial time-saving advantage over manual counting methods.
• * QuantiFly is a valuable, open-source resource for researchers studying Drosophila fitness and behavior.