We have recently published a 3D electron microscopy volume of the whole fruit fly brain. However mapping of synaptic `wiring diagrams' or connectomes of neuronal circuits from this volume is currently completely manual and therefore slow. Our long-term goal is to increase understanding of how circuits process and transform information, both by accelerating connectomics mapping, and increasing the power and accessibility of analysis tools. The overall objectives in this application are to (i) provide newly available whole-brain segmentations to the user community, by extending the widely used, web-based circuit-mapping and analysis platform, CATMAID; and (ii) develop, validate, and apply new analysis tools for delivering insight into circuit structure. These objectives will be achieved through three specific aims: 1) Deploy automatic segmentations to a CATMAID server where members of the fly neurobiology community can create and manage their own collaborative projects; 2) Provide integrated, scalable, and cross-modal software for analysis of connectomes and neurogeometry; and 3) Complete proof-of-principle projects using this infrastructure, including the first whole-brain tractography database, a study of stereotypy in position, number and connectivity of tracts and identified cell types, and a complete map of microconnectivity in the calyx of the mushroom body.
In Aim 1, two recently generated whole-brain segmentations of all neurites and all synaptic clefts will be skeletonized, imported into CATMAID, and provided for use by the fly neurobiologists (>20 labs immediately; the entire community shortly thereafter).
In Aim 2, tools for shape-based neuron search will be integrated into CATMAID, allowing online databases of cell type morphology to assist the reconstruction workflow and find genetic driver lines for behavior and physiology experiments at the bench. Analysis software will be written for automated tractography, location of contralateral circuit element equivalents, and identification of synaptic partners.
In Aim 3, this infrastructure will be used in proof-of-principle analyses designed to test and guide infrastructure development, while providing important biological contributions in their own right. The proposed research is innovative because it will: almost immediately provide segmentation-assisted circuit reconstruction to the fly community in the whole fly brain dataset; provide powerful new analysis tools; and enable individual labs to form and manage their own public or private collaborations in an expertly managed server environment. The proposed research is significant because acceleration of circuit mapping will greatly increase our understanding of information processing in the brain of an important animal model system, the fruit fly Drosophila melanogaster; further, these tools will be applicable to diverse brain tissues and organisms. Ultimately, this infrastructure will therefore contribute to a qualitatively improved understanding of how neuronal circuits process and store information across phyla.
The proposed research is relevant to public health because it provides a new and cutting-edge strategy to accelerate mapping of brain circuits at a synaptic level, a prerequisite to understanding circuit function in normal and diseased brains. The proposed infrastructure is relevant to the part of the BRAIN Initiative's mission that pertains to revolutionizing our understanding of brain function through tool development and improved data analysis technologies.
