Automated, high-throughput quantification of EGFP-expressing neutrophils in zebrafish by machine learning and a...

Area of Science:

• Immunology
• Developmental Biology
• Bioimaging

Background:

• The zebrafish (Danio rerio) is a valuable model organism for studying immune system development and response due to conserved immune cell lineages and transparent larvae.
• Traditional methods for analyzing immune cells in zebrafish larvae are low-throughput, limiting large-scale genetic and chemical screening.
• High-throughput methods are needed to analyze cellular responses in zebrafish for drug discovery and toxicology.

Approach:

• A novel parallelized microscope, the Multi-Camera Array Microscope (MCAM™), was developed for rapid, high-resolution imaging of zebrafish larvae.
• An automated, machine learning-based algorithm was created to segment zebrafish larvae and count fluorescently labeled neutrophils in vivo.
• Image acquisition of a 96-well plate took 75 seconds, with cell counting completed in 5 minutes.

Key Points:

• The MCAM™ system coupled with machine learning enables high-throughput, in vivo quantification of fluorescently labeled neutrophils in zebrafish larvae.
• The automated workflow significantly reduces analysis time from manual methods to minutes.
• The system was validated by comparing algorithmic counts to manual counts, demonstrating accuracy for detecting changes in neutrophil numbers.

Conclusions:

• This optimized method using MCAM™ and machine learning facilitates rapid, statistically significant biological experiments for high-throughput genetic and chemical screens.
• An open-source software package is provided for custom model training, zebrafish localization, and cell count analysis.
• The approach is adaptable for other zebrafish cell lineages and potentially other model organisms.