1032B Poster - 16. Techniques and technology
Friday April 08, 2:00 PM - 4:00 PM
Multiscale, multi-perspective imaging assisted robotic microinjection of Drosophila melanogaster embryos
Authors: Andrew D. Alegria 1; Amey S. Joshi 1; Kunpeng Liu 1; Jorge Blanco Mendana 2; Benjamin Auch 2; Daryl M. Gohl 2,3; Suhasa B. Kodandaramaiah 1,4,5
Affiliations: 1) University of Minnesota Twin-Cities, Department of Mechanical Engineering, Minneapolis, MN; 2) University of Minnesota Twin-Cities, University of Minnesota Genomics Center, Minneapolis, MN; 3) University of Minnesota Twin-Cities, Department of Genetics, Cell Biology, and Development, Minneapolis, MN; 4) University of Minnesota Twin-Cities, Department of Biomedical Engineering, Minneapolis, MN; 5) University of Minnesota Twin-Cities, Department of Neuroscience, Minneapolis, MN
Keywords: q. other (Robotics); q. other (Machine Learning)
Microinjection is an important technique that has enabled transgenesis and targeted mutagenesis in Drosophila, as well as the wide array of methods and genetic toolkits which depend on transgenesis. However, microinjection remains a labor-intensive and highly specialized manual procedure, which makes it a critical bottleneck in the field and thus ripe for automation. Here, we present a computer-guided robot that automates the targeted microinjection of Drosophila melanogaster embryos. The robot uses a series of cameras and microscopes to image a petri-dish containing embryos at multiple magnifications and perspectives. This imaging is combined with a machine learning algorithm and computer vision algorithms to pinpoint a location on the embryo for targeted microinjection with microscale precision. We demonstrate the ability of the robot to successfully microinject Drosophila melanogaster embryos. Results obtained indicate that the robot can significantly increase throughput as compared to manual microinjection since approximately in 1 hour the robot can inject upwards of 200 embryos. Since microinjections are performed in embryos on the petri-dish, the robot further eliminates the preparatory steps such as the dechorionation and placing of the embryos on a glass slide prior to microinjection. Further, the robot is able to maintain survivability rates comparable to that of manual microinjection with survival rates upwards of 50%. We have demonstrated robust and reproducible transposon and PhiC31-mediated transgenesis with automated robotic microinjection of embryos directly on petri-dishes. This microinjection robot provides a means to achieve scalability as additional systems can be built at relatively low cost ($13,000) allowing parallelization of the injection process. In the future, we anticipate the robot can be used to perform high throughput microinjections and potentially enable new types of injection-intensive experiments.