Automated phenotype pattern recognition of zebrafish for high-throughput screening

Area of Science:

• Developmental biology
• Genomics
• High-throughput screening

Background:

• Zebrafish (Danio rerio) are crucial model organisms for genetic and chemical screenings.
• Automating embryo and larvae sorting is essential due to increasing research and experiment volume.
• Manual sorting and feature detection are time-consuming and labor-intensive.

Purpose of the Study:

• To develop an automated platform for image acquisition and classification of zebrafish embryos and larvae.
• To enable high-throughput sorting based on phenotypical features, reducing manual intervention.

Main Methods:

• A prototype platform for automated image acquisition was developed.
• Classification software utilizing image processing and pattern recognition algorithms was created.
• Zebrafish features, including pigmentation, were automatically detected and analyzed.

Main Results:

• The system automatically detects zebrafish embryos, larvae, and their features.
• Zebrafish phenotypes were classified with 79-99% accuracy at 72 hours post-fertilization (hpf).
• The software distinguishes between wild-type and variant phenotypes without user interaction.

Conclusions:

• Automated image analysis and pattern recognition offer an efficient solution for zebrafish sorting.
• This technology significantly enhances throughput and accuracy in zebrafish-based screenings.
• The developed platform and algorithms are valuable for advancing zebrafish research.