Integrated computer vision and microscopy tools for all-optical interrogation of the nervous and muscular systems in Caenorhabditis elegans
Porto, Daniel Akashi
MetadataShow full item record
All animals constantly process sensory information from the environment to guide behavioral responses. A fundamental question in neuroscience is to understand how the nervous system uses neuronal circuits to perform neuronal computations between environmental signals and behavioral outputs, or in terms of systems analysis, what are transfer functions that describe this system’s function? A key challenge in answering this question in humans is the overwhelming complexity of the human brain. Furthermore, technical challenges limit the capabilities and accuracy of measurements of neuronal activity and behavioral outputs while simultaneously probing parts of the circuit. Caenorhabditis elegans is a model organism with a small nervous system consisting of only 302 neurons, and has allowed for the elucidation of genetic and circuit mechanisms of neuronal computations. Additionally, advancements in experimental methodologies such as calcium imaging, optogenetics, and behavior tracking have enabled robust experimentation and analysis. However, conventional instrumentation used to perform these experiments are limited in the integration of these tools and analytical frameworks for extracted data. This thesis focuses on the development of several integrated platforms to perform optical interrogation of various aspects of the nervous system. I use image processing techniques and microscopy tools to improve the throughput and robustness of both experimentation and analysis. These platforms were implemented to address specific biological questions, and were used in experiments involving phenotyping morphology, measuring neuronal activity using functional imaging, activation of excitable cells using optogenetics, and behavior tracking using computer vision techniques. Using these platforms, I developed a reverse correlation methodology to systematically inspect the mechanosensation circuit in C. elegans. I elucidate linear and nonlinear transformations that characterize the temporal and spatial properties of the neural circuitry, providing models that can predict its function. In another application, I used the developed platforms to perform FRET imaging of muscles in freely moving animals. This enabled the investigation of conformational changes of the muscle protein twitchin in vivo. In addition to these two biological investigations, I applied the developed techniques to address other limitations in C. elegans research, including robust calcium imaging analysis in microfluidic devices, as well high-throughput screens for a spinal muscular atrophy model and a lipid storage mutant model. Together, the newly developed platforms and analysis pipelines have improved our ability to investigate activities and functions of neural circuits, and have enabled novel insights about the nervous and muscular systems of C. elegans.