QuantiFly

QuantiFly automates quantification of egg laying in Drosophila melanogaster to provide accurate egg counts for genetic and behavioral studies.

Key Features:

• Automated counting: Uses pattern recognition and machine-learning algorithms to detect and count Drosophila eggs in vial images.
• Bias correction: Implements corrections to an existing algorithm to reduce systematic biases in output.
• Media-specific performance: Achieves average accuracies of 94% on transparent (defined) media and 85% on opaque (yeast-based) media.
• Artifact robustness: Maintains performance in the presence of food surface artifacts such as bubbles and crevices.
• Training on labeled images: Trains models using a labeled subset of eggs from images.
• Batch processing and speed: Performs batch analyses on new images at a rate of a few seconds per image, compared with approximately 40 seconds per vial for manual counting.

Scientific Applications:

• Egg-laying quantification: Quantitative measurement of egg counts in Drosophila melanogaster experiments for genetic and behavioral assays.
• Comparative media analysis: Comparison of egg-laying between transparent (defined) and opaque (yeast-based) media.
• Experimental sensitivity: Detection of experimental differences in egg-laying across conditions given reported accuracy levels.

Methodology:

Applies pattern recognition and machine-learning algorithms trained on a labeled subset of eggs, includes algorithmic bias correction, and performs batch analysis of vial images.