Bryan William Jones - US grants

EY015128 Project Summary: Retinal remodeling This renewal is based on our discovery that retinal degenerations are two separate diseases. Acute primary photoreceptor degenerations turn into chronic neurodegenerations mimicking brain diseases called ?pro- teinopathies?. The neurodegeneration is unremitting, slowly destroying the neural retina with > 90% loss of retinal neurons, severely impeding rescue. The proposal aims to pro?le proteinopathy molecules; identify net- works underlying metabolic collapse; map the nature and scope of rewiring and neuronal loss, and develop a pigmented Tg P347L rabbit model of human adRP. Aim 1. Pro?le neurodegenerative proteinopathy molecules in the neurodegenerative retina. Hypothesis. Retinal neurodegeneration is a proteinopathy. Outcomes: A comprehensive, proteinopathy ?nger- print spanning 6y of disease progression. Signi?cance. Interventions for blinding diseases require survival of the retina. Neurodegeneration must be overcome. Aim 2. Characterize of the metabolic / signaling collapse of the ND retina. Hypothesis. ND corrupts neuronal signaling and energetics. Outcomes: Metabolome status for all neurons and glial cells spanning 6y of ND. Signi?cance. If neurodegenerative retinas cannot sustain activity, therapeutic in- terventions are destined to fail. Metabolic network mapping may reveal druggable targets. Aim 3. Develop a pathoconnectome spanning mid and late ND. Hypothesis. Neural rewiring in the retina is driven by interactions between survivor and degenerating neurons. Outcomes. The survivorship and connectivity of the ND retina. Signi?cance. Different rescue schemas have different targets and the status of each is critical. Aim 4. Develop a pigmented Tg P347L rabbit for retinal degeneration research. Hypothesis. A pigmented eye is optimal for studying of retinal degenerations. Outcomes. A long lifespan, eco- nomically viable, large-eye pigmented model suitable for the FDA Animal Model Quali?cation Program, imag- ing, and analysis of disease progression, rescue and ND interventions. Signi?cance. Developing therapies for human retinal disorders will eventually have to proceed beyond mouse models. A pigmented rabbit is an ideal platform for intervention testing.

Multivariate networks -- datasets that link together entities that are associated with multiple different variables -- are a critical data representation for a range of high-impact problems, from understanding how our bodies work to uncovering how social media influences society. These data representations are a rich and complex reflection of the multifaceted relationships that exist in the world. Reasoning about a problem using a multivariate network allows an analyst to ask questions beyond those about explicit connectivity alone: Do groups of social-media influencers have similar backgrounds or experiences? Do species that co-evolve live in similar climates? What patterns of cell-types support different types of brain functions? Questions like these require understanding patterns and trends about entities with respect to both their attributes and their connectivity, leading to inferences about relationships beyond the initial network structure. As data continues to become an increasingly important driver of scientific discovery, datasets of networks have also become increasingly complex. These networks capture information about relationships between entities as well as attributes of the entities and the connections. Tools used in practice today provide very limited support for reasoning about networks and are also limited in the how users can interact with them. This lack of support leaves analysts and scientists to piece together workflows using separate tools, and significant amounts of programming, especially in the data preparation step. This project aims fill this critical gap in the existing cyber-infrastructure ecosystem for reasoning about multivariate networks by developing MultiNet, a robust, flexible, secure, and sustainable open-source visual analysis system.


MultiNet aims to change the landscape of visual analysis capabilities for reasoning about and analyzing multivariate networks. The web-based tool, along with an underlying plug-in-based framework, will support three core capabilities: (1) interactive, task-driven visualization of both the connectivity and attributes of networks, (2) reshaping the underlying network structure to bring the network into a shape that is well suited to address analysis questions, and (3) leveraging provenance data to support reproducibility, communication, and integration in computational workflows. These capabilities will allow scientists to ask new classes of questions about network datasets, and lead to insights about a wide range of pressing topics. To meet this goal, we will ground the design of MultiNet in four deeply collaborative case studies with domain scientists in biology, neuroscience, sociology, and geology.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Connectomes are Rosetta Stones for discovering how retinas are wired, and revealing how that wiring becomes corrupted in disease. This proposal is crafted around assembly, annotation and analysis of ultrastructural connectomes from normal human retina, providing a normal circuit topology framework to compare against companion human retinal patho-connectomes from human retinitis pigmentosa (RP) and age- related macular degeneration (AMD), hypothesizing that 1) new networks including rod-cone crossover, bipolar cell coupling, nested inhibition, and neurogliovascular (NGV) architectures, structures (cistern synapses, plaques, adherens) are common across mammalians, and 2) seeks to explore how these network topologies are altered in retinal disease. Prior work unmasked unexpected, pervasive complexities in mammalian retina, informing modeling of retinal prosthetics. Speci?c Aim 1. Construction / annotation of a ~100 TB connectome for human (H-RC3). Signi?cance. H-RC3 allows analyses of human cone channels, describing how these networks are blended with existing channels. We hypothesize 1) novel crossover / coupling motifs are present in both midget and diffuse bipolar cells and 2) nested feedback/feedforward architectures are key to midget pathways. With TEM-compliant molecular markers, we will map the heterocellular architecture of the human NGV system in comparison with RC1 and M- RC2. Speci?c Aim 2. Construction / initial annotation of 2 ~50TB connectomes of human RP retina (HRPC1,2- RC4,5). Signi?cance. Annotation/analysis of RC1 revealed unexpected retinal architectures in?uencing all networks: refactored IPL lamination, extensive rod-cone bipolar-amacrine crossover networks, complex homocellular / heterocellular / in-class / cross class coupling, selective bipolar cell loading by ganglion cells. Presuming these networks are present in humans, direct comparisons from SA 1 should establish how networks are altered in human RP, serving as guides to understanding RP progression, and reveal targets for therapeutic intervention. Speci?c Aim 3. Construction / initial annotation of 1 ~80TB connectome of human AMD retina (HRPC3-RC6). Signi?cance. HRPC3-RC6 will allow exploration of altered neuronal, glial and vascular networks in AMD, exploring whether retinal remodeling and plasticity are similar for different retinal disease mechanisms. We now know retinal remodeling occurs in AMD, and that GABAergic amacrine cell networks are involved. We do not know if the rules behind retinal network alterations are similar across disease pro?les such as RP. Early evidence suggests photoreceptor degeneration, remodeling and progressive neural atrophy are separate processes. Testing this in non-RP models will guide understanding and create a complete connectome database of AMD retina that includes the choroid, neural and glial components.

Project Summary-Imaging Core The OVERALL aims of this Vision Research Core (VRC) are to provide: · access to resources outside the scope of individual R01 awards · access to technical expertise outside the scope a single laboratory · staff training to remove barriers to efficient translational research and collaboration · collaboration initiatives among VRC labs The research areas supported by the VRC span the analysis of fundamental biology of normal tissues involved in the visual system as well as a range of cutting-edge basic science initiatives involved in treatments of retinal degenerations, developmental disorders, glaucoma and other disorders. We have implemented four resource modules that continue the natural evolution of how this research group works together, serving 17 investigators holding 22 NEI R01 awards. The Imaging Module provides a range of imaging (TEM, confocal, metabolomic CMP, scanning optical) and computing services (imaging, database, mathematics) based on strengths of core laboratories and the tradition of excellence of the UU School of Computing, whose descendants founded Adobe Systems, Silicon Graphics, Netscape and Pixar. Collaborations among these groups have transformed software tools for TEM and confocal imaging. Specifically, the Imaging Module provides: ? novel, powerful high-speed automated TEM for VRC investigators, regardless of experience, that delivers top quality TEM imagery in a readily navigable format using the Viking web-application. ? metabolic mapping onto anatomy using Computational Molecular Phenotyping (CMP) resources. ? research grade confocal resources and management both in the JMEC proper and the JMEC vivarium; expanded confocal resources are planned. ? high-speed optical scanning microscopy for high-throughput tissue / immunocytochemical analysis. ? extensive imaging, database and mathematics software holdings and training expertise ? large scale data storage for the VRC (up to 0.5 petabyte).