The zebrafish (Danio rerio) has emerged as a robust model organism owing to its transparency, fecundity and genetic capabilities including transgenesis and gene knock-out. In recent years, zebrafish behavioral phenotyping has been increasingly used for studying toxicological and teratogenic effects, neurological disorders and certain diseases. However, to fully exploit the advantages of the zebrafish model novel sample handling, imaging and data analysis and storage platforms are required. The Screening Facility linked to the European Zebrafish Resource Center (EZRC) has developed high-throughput behavioral screening platforms to systematically characterize zebrafish embryonic behavior. The “zebrafish sorting robot” automatically pipettes single or multiple zebrafish eggs or embryos into standard microtiter plates. This robot can operate in complete darkness and thereby facilitates experiments that demand the absence of ambient light. The “Droplet-Microarray Sandwiching Chip” is an array of superhydrophilic spots separated by superhydrophobic borders. By rolling a suspension of zebrafish embryos on the surface of the chip using a pipette we are able to spread zebrafish embryos in one-embryo-per-droplet manner without need for pipetting. This device has been successfully used to screen compound libraries and performs comparably to standard microtiter plates. Moroever, we developed a high-throughput Photomotor Response (PMR) imaging system that measures the response of zebrafish embryos at 30 – 40 hours post fertilization to flashes of bright light. This PMR platform can be completely remotely controlled thereby ensuring no human interference and extraneous sound and vibrations. Another behavioral platform quantifies the startle response using an automated acoustic vibration system combined with a very high frame rate high-resolution camera. Finally, a robotic imaging platform was developed that can assess long term locomotion and feeding behavior in an unsupervised way. In addition to the development of the imaging systems, specialized real-time and quantitative image analysis and processing pipelines have been established. These high-performance computing pipelines extract locomotor kinematics and behavioral dynamics. Furthermore, since behavioral experiments are data intensive (often leading to 7 – 10 tera bytes of data per experiment per day), a dedicated large scale data facility that has more than several peta bytes of storage and archival capacity has been established.
References: Marcato, D., et al. An Automated and High-throughput Photomotor Response Platform for Chemical Screens Proc., 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS), 2015.