Genetic Approaches to Dissecting Neural Computation in the Visual System. Tom Clandinin. Dept of Neurobiology, Stanford University, Stanford, CA.

   Our understanding of the complex neural circuits that underlie most visual behaviors is extremely limited. Forward genetic approaches have contributed considerably to our understanding of many aspects of biology; our goal is to adapt analogous methods to the functional dissection of visual circuitry. These methods, then, provide an entry point to examining the computational roles of specific neurons. We developed a convertible enhancer trap, the InSite system, which allows rapid replacement of genetic effectors, enabling expression patterns to be refined through intersectional approaches, and facilitating independent manipulation of gene expression in multiple cell types simultaneously. We have also developed both high-throughput, as well as single fly paradigms to examine behavioral responses to distinct visual cues, and used these in forward genetic screens to identify cell types that play critical roles in visual responses to motion and polarized light. By then imaging calcium responses in these cells, and by measuring their electrophysiological responses, it becomes possible to relate particular visual stimuli to specific neural responses, to computation and behavior. These studies have revealed deep similarities in circuit architecture and computational strategy between flies and vertebrates.