Megabouts · AI-Enhanced Zebrafish Screening: A Low-Cost Solution for High-Throughput Biomedical Research

Thorough, quantitative analysis of behaviour is essential for advancing neuroscience and neurological disorder research and has the potential to significantly improve the characterization of phenotypes in genetic, pharmacological and toxicological screens commonly employed in the pharmaceutical and biotech industries. Achieving high-throughput screening, without compromising the richness of phenotypic data requires scalable, standardized and cost-effective solutions for behavioral recording that also allow seamless integration across labs. To address this need, we developed megabouts.ai (Jouary et al., Biorxiv, 2024) an AI-based pipeline for behavioral analysis capable of reliably inferring high-speed behavioral kinematics from diverse and low-resolution data sources. Complementing this, we have designed a low-cost, modular behavioral testing system optimized to leverage the strengths of the megabouts.ai pipeline, and enable sophisticated behavioral experiments compatible with high-throughput screening approaches. In this proposal, we will develop and validate these tools across a range of behavioral screening paradigms, exploring their potential applications in both basic and translational research. Our goal is to disseminate these innovations, promoting their effective use in academic and industry settings, unlocking new experimental possibilities and enhancing the capabilities of existing commercial and custom-built solutions.