Duke University

NSF Awards · FY 2025 · 2025-01

There are more lakes and ponds in the Arctic than anywhere else in the world. These lakes provide habitat for wildlife and support the subsistence activities of Indigenous communities. Arctic lakes are also an important component of the climate system. Changes in lake area could strengthen or reduce climate change feedbacks, depending on whether lakes expand or shrink. Previous research shows that climate change is causing Arctic lake area to change, but the overall trend is unclear. This project uses high-resolution satellite data to assess the direction and magnitude of surface water change in the Arctic. Based on project findings, the PI will develop a set of best practices for tracking Arctic lake area change and determine whether lakes are expanding or shrinking in five Arctic regions. This research aims to reduce uncertainties in the magnitude and direction of Arctic lake area change by advancing understanding of interannual variation and developing a methodological approach to identify long-term trends in Arctic surface water. The researcher will use Planet (3 m) images to generate multi-year high resolution maps of surface water in five Arctic regions: the Arctic Coastal Plain, the Yukon-Kuskokwim Delta, the Yukon Flats, the Mackenzie River Delta, and the Tuktoyaktuk Peninsula. These high-resolution maps will be used as a reference for quantifying the accuracy and sensitivity of existing coarse resolution (i.e., Landsat and MODIS) surface water products. Based on the outcome of this methodological comparison, the researcher will develop a set of best practices for tracking Arctic lake area change with coarse resolution (e.g., Landsat) images, and, in regions where previous work is inconclusive, determine whether lake area has increased or decreased over the past twenty years. 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.

NIH Research Projects · FY 2025 · 2025-01

Project Summary The cortex is responsible for many of our uniquely human cognitive abilities and distinguishes our closest evolutionary relatives. Humans diverged from chimpanzees ~7mya and have accumulated several genomic differences that could help explain human-specific aspects of cortical development. Many of the genomic differences between humans and chimpanzees are in non-coding elements where they have been traditionally studied as regulators of gene expression. In this proposal, I provide preliminary evidence that indicates additional roles for these human-specific sequences in post-transcriptional gene expression. Over 40% of these human- specific sequences are found within introns where I hypothesize, they contribute to the regulation of splicing during human-specific neurodevelopment. In Aim 1, I will test the function of intronic human-specific sequences to enhance and repress splice sites and ultimately alter species-specific isoform expression. Next as a proof of concept, in Aim 2 I will investigate the consequences of changes in splicing and isoform production in neurodevelopment. Specifically, I will be testing changes in splicing associated with an Autism-associated SNP in the intron of Tenascin C (TNC). The SNP is located in a human-specific sequence and reverts the intronic sequence back towards the chimpanzee sequence. The proposed research will define a novel mechanism of action for human-specific non-coding regions of the genome in the regulation of post-transcriptional RNA processing during cortical development and disease.

NIH Research Projects · FY 2026 · 2025-01

Attention-deficit/hyperactivity disorder (ADHD) is associated with substantial societal burden and personal impairment across the lifespan. Although empirically-based treatments exist, ADHD is highly heterogeneous with respect to treatment response as well as ADHD-related clinical profiles, including psychiatric comorbidity, cognitive performance, and sluggish cognitive tempo (SCT). Personalized treatments attuned to the biology and behavior of the child are critically needed to reduce the burden of ADHD. Sleep disturbances, particularly difficulties falling asleep and delayed/irregular sleep schedules, are implicated in ADHD and may represent key candidates for elucidating clinical heterogeneity and personalizing ADHD treatment. Two distinct mechanistic pathways (i.e., behavioral and biological) may contribute to sleep disturbance in ADHD. Both are characterized by difficulties falling asleep and delayed/irregular sleep schedules but are mechanistically distinct. In the behavioral pathway, sleep disturbances stem from ADHD symptoms (e.g., disorganization) and psychiatric comorbidity (e.g., oppositionality), while the biological pathway is characterized by an endogenous circadian rhythm delay. Although the behavioral pathway has received significant scientific attention to date, very little is known about the prevalence, impact, and optimal treatment of the biological (circadian) pathway in middle childhood—the peak period of ADHD diagnosis. Importantly, the development of identification and treatment strategies for children with circadian dysfunction has been limited by gold-standard circadian assessment (dim light melatonin onset [DLMO]), which is burdensome for families and difficult to implement in routine clinical settings. The overarching objective of this study is to characterize circadian function among 250 children with ADHD (ages 6-9) using gold-standard DLMO assessment and to establish the implications of circadian dysfunction for clinical and cognitive functioning. In addition, this study employs innovative wearable sensors (e.g., heat flux, rest/activity, mattress-based) to determine whether low-burden, in-home assessments have utility in identifying circadian dysfunction among children with ADHD. Aim 1 will characterize individual differences in circadian function, including behavioral and biological sleep profiles, in pediatric ADHD using lab- based DLMO, actigraphy, and parent report. Aim 2 will examine associations between circadian dysfunction, ADHD clinical presentation, and cognition. Aim 3 will investigate the utility of wearable sensors for indexing circadian dysfunction in children with ADHD. If successful, this study may transform clinical practice by facilitating holistic, objective sleep assessments in pediatric settings, which may spur the development of personalized treatments (e.g., behavioral sleep management therapy for behavioral pathway, chronotherapy for the biological pathway). Although we focus on ADHD, our methods have transdiagnostic applications. From a public health perspective, the improved measurement and treatment of sleep problems has the potential to improve educational, interpersonal, and occupational outcomes and result in broad economic savings.

NIH Research Projects · FY 2026 · 2025-01

Project Summary/Abstract Metastasis is the leading cause of cancer-related deaths. In patients with metastatic castration-resistant prostate cancer (mCRPC), although bone is the most prevalent site of metastasis, the existence of metastases in visceral organs such as the liver and lungs leads to the poorest prognosis. Current androgen receptor signaling-targeted therapies have not vastly improved overall survival in mCRPC patients. There remains an urgent need for the development of novel therapies for mCRPC patients, particularly those visceral and bone mCRPC patients who have the worst prognosis or a high prevalence. We and others have found that mCRPC cells highly express the oncogenic transcription factor (TF) HoxB13, which has been shown to promote CRPC growth, invasion and metastasis, However, the precise molecular and genomic mechanisms that underlie the oncogenic functions of HoxB13 remain largely unknown. Furthermore, like the majority of TFs, HoxB13 is considered untargetable by traditional, small molecule-based drug design. Gene therapy is a critical alternative strategy with the potential to directly target such traditionally undruggable genes. We have developed a lipid nanoparticle (LNP) system for safely delivery of the highly specific RNA-targeting CRISPR/Cas13d (CasRx) system to knock down mRNAs of transcription regulators in prostate cancer. By combing the LNP-CasRx system with the visceral organ-targeting SORT (selective organ targeting) nanotechnology, or incorporating the bone-targeting chemical alendronate into the LNP, and further modifying these visceral- or bone-targeting LNP surfaces with the prostate cancer-specific aptamer E3, we have successfully developed and tested a tri- targeting LNP system, simultaneously targeting 1) specific organs including liver, lungs or bone; 2) metastatic cancerous cells (mCRPC cells); and 3) mRNAs encoding TFs (HoxB13). In this proposal, we hypothesize that HoxB13-driven oncogenic transcription and CRPC metastasis can be counteracted by a CasRx-based LNP system with specificity, efficacy, and safety. In Aim 1, we will uncover molecular and genomic mechanisms by which HoxB13 activates oncogenic transcription in CRPC. In Aim 2, we will assess the therapeutic efficacy, safety and molecular impact of the CasRx-based tri-targeting LNP system in CRPC. The data generated from this proposal will provide key preclinical information about the efficacy and safety of this tri-targeting therapeutic strategy, forming a basis for future medical application of a nanoparticle-delivered RNA targeting system to inhibit HoxB13-driven oncogenic transcription in metastatic CRPC.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT A fundamental goal in neuroscience is to determine how information is coded by the spiking activity of a neural population. In the retina, parallel circuits process different features of visual input into a pattern of spatiotemporal activity across retinal ganglion cells (RGCs). Each RGC type corresponds to a distinct parallel processing channel, collectively aimed at simultaneously encoding diverse features from the entire visual field. Given the limited capacity of the optic nerve for information transmission, the RGC population has evolved to be highly efficient at this task. A diverse set of complex linear and nonlinear computations achieves this efficient coding, but the parallel retinal circuits supporting these computations are poorly understood. This project will investigate how feed-forward inhibition onto RGCs contributes to the parallel processing of visual scenes. In particular, I will determine how GABAergic inhibition shapes the feature selectively of RGCs and the signaling performed by a population of neurons. To identify the role of feed-forward GABAergic inhibition onto different RGC types, I have developed an innovative approach that combines a chemogenetic technique called DART to selectively modulate GABAA receptors on RGCs with simultaneous recording of spiking activity from hundreds of RGCs with multi-electrode arrays. In Aim 1, I will evaluate how inhibition shapes the RGC- type specific spatiotemporal linear filter, which describes how visual input is encoded linearly in space and time. In Aim 2, I will identify how feed-forward inhibition contributes to static nonlinearity, which defines the relative sensitivity of RGC responses to different visual stimuli. Lastly, in Aim 3, I will determine how GABA inhibition shapes correlated activity, a crucial aspect for population-level decoding. The expected outcome of this proposal will be identifying how synapse-level mechanisms shape the output of the retina. The project is significant because determining the inhibitory signaling pathways underlying neural computations is critical for comprehending information processing with broader implications for fields like perception, behavior, and adaptation.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT: Virtually every fundamentally important cellular process, including cell migration, division, death, differentiation, and metabolism, has been found to be dependent on mechanical cues. Similarly, many disease states impacting human health, but lacking effective treatments, are at least partially caused or exacerbated by perturbed mechanical cues. Examples include cancer and fibrotic disease, which are associated with increased stiffness of tissues, and atherosclerosis, which is associated with perturbed hemodynamics. Despite the clear importance, the underlying molecular mechanisms through which cells sense and respond to mechanical cues, commonly referred to as mechanosensitivity, are poorly understood. In particular, the lack of knowledge regarding both the key mediators and the pertinent regulatory mechanisms highlights a substantial knowledge gap associated with the inability to translate the current understanding of mechanosensitivity into the detailed mechanistic knowledge required for manipulating cell behavior and developing novel therapeutics. These deficits are driven by a significant technological gap. The current challenge is that most existing techniques in cell biology assume a purely biochemical basis for biological function and either destroy or cannot distinguish these mechanical interactions. In particular, there are relatively few technologies to assess how the application of force affects protein functions in cellulo. Our work is guided by a new conceptualization of how the specificity of mechanosensitivity is determined by the mechanical state, defined by localization, ability to form interactions with other proteins, degree of mechanical loading, and amount of phosphorylation of the pertinent linker protein. The proposal will address three goals. First, leveraging our expertise in protein engineering, we will create and use a suite of biosensors to elucidate how phosphorylation affects other aspects of the mechanical state of linker proteins, rigidity sensing, and force transmission between the cell and the ECM (Goal 1). Second, building on previous work studying force-sensitive protein interactions, we will develop a novel approach for unbiasedly identifying the key mediators of mechanosensitivity for a given mechanical linker protein (Goal 2). Third, we will evaluate the role of these mediators in stiffness sensing and various forms of cell migration, potentially identifying proteins that drive mechanosensitive cellular processes and disease states (Goal 3). This R35 will give the lab the flexibility and power to advance the molecular understanding of mechanosensitivity and open new avenues for the study of mechanosensitive cellular processes. Findings from the proposed studies will also lay the foundation for advances in the diagnosis, treatment, and prevention of mechanosensitive diseases.

NSF Awards · FY 2025 · 2025-01

The project will test how fire affects biodiversity in fire-prone ecosystems. Disturbance is important in maintaining species diversity in most ecosystems. Fire is a common disturbance that occurs in many systems such as longleaf pine savannas. Although research has tested the effects of fire on biodiversity, little is known about how climate change might alter these effects. For example, longer or more intense droughts might reduce population recovery after a fire. This project will test how fire frequency affects biodiversity and how climate change might modify fire effects in the future. Conservation of threatened ecosystems such as longleaf pine savannas requires the ability to predict how populations will change in the future. Longleaf pine savannas are home to many threatened plants and animals like the Venus flytrap and red-cockaded woodpecker. This project will provide recommendations for prescribed burns in longleaf pine savannas and will create a web-based tool for land managers in the southeastern U.S. that will help them make conservation plans. Additionally, the project will train graduate and undergraduate students in ecology and conservation. The project will explicitly test predictions that the optimal fire management strategy for Venus flytraps will differ in a future climate. These predictions were constructed using data collected under ambient variation in climate and fire regimes, rather than extreme values of climate consistent with future climate change, or experimental manipulations of fire that disentangle correlations between historical fire frequency and current fire effects. Using demographic data on Venus flytraps collected from almost-factorial manipulations of fire frequency, drought, and warming conducted across a broad geographic area, the PIs aim to construct a climate- and fire-driven integral projection model that explicitly includes site-specific effects. We will validate the model using independently collected abundance data on fire and climate effects. The proposed work will also estimate the degree to which Venus flytraps can be used as an indicator species, where a good indicator is one that accurately predicts changes in abundance of other species in response to fire and climate, rather than simply the presence of other species or of high levels of biodiversity. Assessing the indicator potential of Venus flytraps will help conservation managers identify fire frequencies that could bolster biodiversity or abundances of species of concern. These efforts will culminate in generalizable insights underlying disturbance management in a future climate and the development of a framework for assessing the utility of indicator species. This project is jointly funded by the Divisions of Environmental Biology and Integrative Organismal Systems through the Partnership to Advance Conservation Science and Practice Program. 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.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT: Tolerance to Allogeneic Hearts via Implantation of Cultured Donor Thymus Advances in immunosuppression have improved patient outcomes after allogeneic heart transplantation, but patients remain at risk for life-threatening infection, chronic rejection, and end-organ failure. Development of tolerance to the transplanted heart would reduce or eliminate the need for continuing immuno-suppression, which should improve both allograft and patient survival. Our goal is to use donor thymus tissue to induce central tolerance to a heart transplanted from the same donor. We have shown that cultured allogeneic thymus tissue implantation (CTTI) into athymic children can establish a fully functional immune system that is tolerant to MHC antigens present on the donor thymus as well as on recipient tissues. We showed that CTTI can also induce robust donor-specific tolerance in immunocompetent rat recipients of allogeneic hearts matched to the thymus donor. This project will extend these exciting and potentially paradigm-changing discoveries to an immunocompetent large animal model where clinical safety and efficacy of tolerance induction via CTTI can be assessed in preparation for human clinical trials. The specific aims are: 1) To assess the safety and efficacy of cultured thymus tissue implantation (CTTI) for induction of tolerance to co-transplanted same-donor cardiac allografts in young miniature swine; 2) To determine the efficacy of CTTI using thymus from young adult donors for immune reconstitution and tolerance development in miniature swine; and 3) To develop evidence-based inclusion criteria for expanding the human thymus donor pool across the age range of potential allograft donors. These studies leverage our unparalleled expertise in human thymus biology, clinical CTTI, and heart transplantation to induce central tolerance to transplanted hearts. Results will be applicable to recipients of all types of allografts and should provide additive or synergistic benefits when combined with tolerance-inducing therapies that act on cells in the periphery. Together, these studies will advance toward the long sought-after goal of generating the robust tolerance of transplanted organs that is needed to markedly improve the survival of allografts and recipients.

NSF Awards · FY 2025 · 2025-01

Ushering in a new era of spectrum sharing requires dynamic spectrum access (DSA) that natively supports both primary and legacy users, while creating new opportunities for spectrum utilization. A comprehensive blend of technical, economic, and policy-based solutions is required to realize this vision, including potential modification to existing cellular standards to ensure that future 6G standards are inherently “sharing native”. Precise, low-latency, and localized spectrum usage monitoring that is aware of and integrated with the cellular Physical (PHY) and upper layers in the networking stack is essential for facilitating effective spectrum utilization and sharing in Spectrum Era 4. However, existing spectrum sharing systems typically rely on a separate monitoring network comprising dedicated, costly, and sparsely deployed spectrum sensors, e.g., the Citizens Broadband Radio Service (CBRS) networks rely on an environmental sensing capability (ESC) sensor network deployed in coastal areas to detect transmissions from Navy vessels and radars. This project aims to realize a transformative vision for spectrum sensing in Spectrum Era 4, which supports dense and in-situ spectrum sensing with significantly enhanced sensing resolution across the temporal and spatial domains, improved energy efficiency, and cooperative sensing strategies that are aware of the cellular protocols. As such, it has the potential to revolutionize the next generation of cellular technologies (e.g., 6G and beyond) to be sharing native with significantly enhanced spectrum awareness and sensing resolution. This project targets the following scientific contributions from three interdisciplinary and interrelated research thrusts. (i) Development of ultra-efficient, single-shot, analog cross-correlators (X-Corr) capable of computing the cross-correlations between input signals and template waveforms across varying lags, enabling spectrum sensing with ultra-low latency. Using the margin computing paradigm, analog X-Corr with superior energy efficiency and (>1,000 TOPS/W) can be designed and realized in integrated circuit (IC) implementations without compromising the computing speed or precision. (ii) Design of protocol-aware configuration and adaption for X-Corr to enable fine-grained, in-band spectrum sensing. This allows for detailed sensing of spectrum occupancy and detection of interference signals at the symbol or slot level (a few to 10s of microseconds) with both known and unknown features (e.g., for airborne and ground radars) and employ diverse PHY layers (e.g., 5G New Radio and Wi-Fi). (iii) Optimized deployment and configuration of a network of densely deployed X-Corr sensors to facilitate cooperative, in-situ spectrum sensing that is aware of the communication standards. Such a network also enhances the ability to localize and track interference sources with significantly lower latency and cost. Evaluation of the proposed research includes analysis, simulations, IC implementations, circuits-system co-design and integration, as well as field experiments using local and community wireless testbeds. 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.

NSF Awards · FY 2025 · 2025-01

The broader impact/commercial potential of this Partnerships for Innovation – Research Partnerships (PFI-RP) project is to disseminate a transformative three-dimensional (3D) ultrasound imaging technology that provides an accessible and practical solution for comprehensive tissue examination in clinics. The technology has the potential to address technical limitations of 2D imaging and make ultrasound a more reliable tool for early detection, diagnosis, and prognosis of many diseases that disproportionally impact people from low-income and under-resourced communities. The technique also provides the research community with new 3D imaging capabilities to probe the structure and function of deep tissues noninvasively and holistically in vivo, facilitating new discoveries and development of new therapies. The approach equips 2D ultrasound imaging systems with the ability to perform 3D imaging, results in lower costs and increased accessibility, making broad and direct impact on patient care globally. In addition, the technology will provide a viable 3D imaging solution for the ultraportable, pocket ultrasound devices that are becoming increasingly popular. Since many of these devices are not used by trained sonographers, 3D imaging is particularly useful for mitigating operator dependence. As such, the technology becomes essential because it will conveniently enable 3D imaging for the ultraportable at-home ultrasound market. The proposed project presents a highly practical and cost-effective solution to democratize ultrafast 3D ultrasound imaging because it instantly converts existing 2D ultrafast ultrasound imaging systems into 3D-capable devices. The technology addresses the market needs of 3D imaging by working with existing ultrasound systems and probes, lowering financial barriers for new users, breaking existing technical barriers of 3D ultrasound imaging, and enabling advanced imaging modalities that are not possible with conventional techniques. The research objective of this PFI-RP project is to conduct use-inspired research to overcome the knowledge gaps and technical challenges of the 3D imaging technique, making this technology ready to be transferred to the commercial ultrasound market. The proposed project will improve various components of devices to improve imaging quality. The proposal will also develop automated calibration and synchronization methods to facilitate streamlined user experience. Designated control hardware and software interface will be developed for integration with commercial ultrasound systems. Finally, in vivo animal imaging studies will be conducted to evaluate the imaging performance of the new 3D imaging device. Successful completion of this project will culminate in a commercially viable 3D imaging technology that is poised for immediate commercialization. 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.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT Using recombinant antibodies targeting common oncogenic mutations, we have recently demonstrated that dimeric IgA penetrates human epithelial cancer cells through PIGR-dependent transcytosis, neutralizing mutated oncodrivers and expelling them outside the tumor cell. Accordingly, targeting KRASG12D with dimeric IgA abrogated the growth of different carcinomas. Moving forward, our primary objective is to translate tumor cell-penetrating Abs into the clinic. However, challenges arise due to the low yield associated with dimeric IgA production and the comparatively extended half-life of IgG antibodies. To expedite clinical translation and maximize therapeutic effectiveness, we have engineered new IgG4 antibodies incorporating a PIGR-binding peptide at the C-terminus of the Fc domain. PIGR-binding IgGs penetrate cancer cells and elicit Ag-specific control of tumor growth. Our central hypothesis is that our mutant KRAS-specific recombinant IgGs control KRAS-driven carcinomas that became resistant to small molecule KRAS inhibitors, with longer half-lives and production yields than recombinant dIgA, while making tumor cells sensitive to T cell-mediated killing. This hypothesis will be tested in these Aims: In Specific Aim 1, we will elucidate the mechanisms governing the effectiveness of our new PIGR-binding mutation-specific IgGs. Besides understanding how these Abs neutralize antigenic mutations inside tumor cells, we will select a candidate for clinical development, based on effectiveness, the capacity to target multiple mutations, and production yields. In Specific Aim 2, we will define how transcytosis of PIGR-binding IgG impacts T cell- dependent anti-tumor immunity. We will define mechanistically-driven and safe interventions that synergize with PIGR-binding Abs, possibly including immune checkpoint inhibitors. In Specific Aim 3, we will determine whether tumor cell-penetrating Abs overcome resistance to inhibitors across various mutations. We will establish a range of human cancers sensitive to our new immunotherapy, including tumors that became resistant to inhibitors. These studies will not only bridge existing knowledge gaps regarding transcytosing antibodies, but also pave the way for a new generation of immunotherapies that effectively combat aggressive human cancers driven by mutational hotspots, and perhaps other cytosolic proteins.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT The objective of this proposal is to determine how cardiac and skeletal muscle myocytes, mitochondrially dense and highly oxidative cells, manage succinyl-CoA (SucCoA). SucCoA serves three conditionally essential metabolic fates: forward tricarboxylic acid cycle (TCAC) flux, heme synthesis, and ketone oxidation. Indeed, sucCoA is a limiting metabolite for the rate-determining enzymes in heme synthesis and ketone oxidation, Delta-aminolevulinate synthase 1 (ALAS1) and Succinyl-CoA:3-ketoacid CoA transferase (SCOT), respectively. In myocytes ketones are a major fasting energy source and heme is the primary oxygen carrier in the form of myoglobin. Heme is also a cofactor for multiple proteins, including several in the electron transport system, rendering it essential for mitochondrial biogenesis. We identified a nutrient dependent reciprocal regulation of ALAS1 and SCOT, that we hypothesize serves to prevent dual pulls on the sucCoA pool. Interestingly, this regulation is lost in heart failure (HF) and sucCoA is depleted. HF is a disease characterized by a progressive deterioration in cardiac pumping capacity concomitant with alterations to oxidative metabolism, including a stark increase in reliance on ketone oxidation at the level of SCOT and β- hydroxybutyrate dehydrogenase 1 (BDH1). While increased ketone oxidation is an emerging hallmark of HF, far less is known regarding heme or the intersection of heme and ketone metabolism in HF. We hypothesize that the unique demands of hypertrophic and failing hearts, which grow increasingly dependent on both ketone oxidation and heme biosynthesis, result in a dual pull on the sucCoA pool, thereby limiting substrate flux through both SCOT and ALAS1. In accordance, we propose two aims to test our hypotheses that: 1) sucCoA availability for heme biosynthesis in myocytes depends on carbon source, nutritional status, and ALAS1 interactome, and 2) that sucCoA availability limits heme synthesis and mitochondrial biogenesis in hypertrophied and failing hearts. To test this model, we will combine state-of-the-art molecular profiling tools, Langendorff perfused hearts, and metabolic flux analysis with stable isotope tracers, ultimately filling fundamental gaps in knowledge of myocyte heme biology and furthering our understanding of HF pathophysiology. These studies will employ primary human myocytes for flux and interactome analysis and a novel myocyte specific double SCOT and BDH1 knockout mouse (MCK-DKO). The training environment at DMPI is uniquely suited to support these studies due to its robust core facilities and trainee support. In fulfilling the aims proposed herein, the applicant will simultaneously expand their technical skillset and integrate their computational and basic science skills. This supports the applicant’s long-term goal of becoming an academic investigator. The proposed work will expand our fundamental understanding of heme biology and cardiovascular physiology, ultimately contributing to new therapeutic approaches for HF.

NSF Awards · FY 2025 · 2025-01

The brain is the most complex organ in the human body. How does the brain work remain one of the most challenging scientific problems for humanity. For decades, scientists and engineers continually develop and refine new methods and techniques to advance our understanding of the brain. Out of these tools, imaging is essential for deciphering the brain because it allows us to directly visualize and investigate the complex brain tissues and their organizations and functional networks. However, even with the multitude of brain imaging technologies that are currently available, our ability to probe deep brain tissues beyond the cerebral cortex is still limited. This limitation can largely be attributed to the physics of imaging, which dictate the inevitable trade-off between how small of an object we can see and from how deep we can see them. This shortcoming ultimately limits our ability to explore beyond the superficial tissues of the brain and to understand how the human brain works in its entirety. The long-term objective of this CAREER proposal, therefore, is to overcome this shortcoming by developing a new ultrasound imaging technology that can probe deep brain functional neural activities at a microscopic spatial resolution. Our technique leverages the power of deep learning and ultrafast ultrasound imaging to break the barrier of imaging speed for conventional super-resolution ultrasound. If successful, this transformative new technology will become a paradigm-shifting imaging tool that provides functional brain mapping at a much finer spatial resolution with a much deeper and wider territory than ever before. The unique capabilities of this new imaging technology will also open new doors for many under-explored opportunities in both basic neuroscience research and in many neurological disease applications. The goal of this CAREER proposal is to develop a new and transformative functional brain imaging technology that allows continuous, real-time monitoring of neural activities of the entire brain at a micron-scale through intact skull. Thrust 1 will focus on improving the temporal resolution of conventional super-resolution ultrasound imaging by developing deep learning-based super-resolution imaging techniques. Thrust 2 will address the computational challenges associated with ultrasound image reconstruction by developing a new ultrafast ultrasound system based on modern high-speed FPGAs. Thrust 3 will concentrate on developing phase aberration correction methods based on deep learning and novel 3D ultrafast imaging techniques to achieve robust intact skull imaging of the whole brain. In vivo mouse brain imaging studies will be conducted throughout the technical thrusts to evaluate and validate the performance of the newly developed super-resolution imaging techniques. If successful, the proposed work will result in a new, radiation-free, low-cost, and widely accessible functional brain imaging technique that will be the first to enable noninvasive probing of in vivo, deep-brain neural activities with high spatiotemporal resolution. In addition to the technical thrusts, this CAREER proposal also includes educational and outreach programs aimed to instill in the new generation of students the desire to improve the standard of healthcare such that all patients have access to state-of-the-art treatment, diagnostic, and screening options. By providing research opportunities, creating ultrasound engineering labs, developing innovative teaching strategies, and establishing new courses, this CAREER proposal will provide these students with the knowledge and tools necessary to create actionable changes within the opportunities presented. 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.

NSF Awards · FY 2025 · 2025-01

Humans are part of a rich global ecosystem containing vast numbers of different species of plants and animals. These highly diverse species interact in complex ways to fundamentally impact human life, health, and economy. Remarkably, most species on earth remain unknown to science. This is particularly true for small organisms, such as insects and fungi, that operate behind the scenes but have a fundamental impact on breaking down and recycling waste and on food production. Particularly given these critical roles, there is growing concern that global environmental change may lead to a reduction in biodiversity and impact which species occur where and how they interact. To reliably study such processes and obtain critical data for making policy decisions, it is imperative that automated methods are available for biodiversity monitoring. This project develops transformative new wireless communication and artificial intelligence (AI) tools for automating biodiversity data collection, processing, and analysis. These tools can be used to infer which bird, insect and bat species are present at any given location and time based on their vocalizations, while assessing relationships with factors ranging from climate to urbanization. The AI tools developed by this project will impact not just biodiversity research but many other fields across the sciences and industry. Young researchers involved in the project will receive training in developing and applying new AI methods in challenging settings. This project develops novel AI methods and applies wireless communication technologies for biomonitoring. Under the ongoing biodiversity challenges, methods are needed for autonomously inferring which species are present in different spatial locations; such monitoring provides critical data for studying impacts of climate change and environmental disruption on biological community dynamics. Current data on biodiversity are heavily taxonomically and spatially biased. Scientists still know remarkably little about the dynamics of species communities, driving which species are present, the interactions among these species, and the impact of disruptions. There have been recent improvements in labor-intensive methods for biodiversity monitoring across broad groups of taxa, including birds, mammals, insects, and fungi. However, data collection remains very costly and requires substantial human intervention and management. Contemporary tools for species classification are error-prone; failing to account for these errors can lead to inaccurate scientific conclusions and flawed policy recommendations. Next-generation biomonitoring requires cost-efficient technologies for automatic, adaptive sampling and flexible inference to characterize and learn from ecological conditions in real time. This project develops transformative new tools for fundamentally improving biodiversity monitoring having immense societal and scientific impact. Fundamental innovations include: (1) probabilistic paradigms to account for errors in inferring species composition from audio, imaging and DNA-barcoding data; (2) new classes of interpretable and identifiable AI-Joint Species Distribution models to characterize how community composition is driven by biotic and abiotic factors; (3) improvements in wireless communication technologies for remote monitoring; and (4) adaptive designs to optimize allocation of limited resources to maximize learning of ecological community dynamics. 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.

NIH Research Projects · FY 2026 · 2025-01

ABSTRACT Heparin-induced thrombocytopenia (HIT) is a thrombotic disorder caused by ultra-large immune complexes (ULICs) composed of IgG antibodies (Abs) to platelet factor 4 (PF4) and heparin (H). Even with prompt recognition, discontinuation of heparin, and institution of alternative anticoagulants, thrombotic risk remains high and persists for weeks. Studies from our last funding cycle suggest that therapy targeting complement may improve outcomes. We showed that HIT ULICs potently activate the classical pathway of complement, elicit complement-dependent binding of ICs to neutrophils (PMNs) and monocytes, and promote Fcγ receptor (FcγR)- dependent PMN degranulation and monocyte tissue factor (TF) expression. Importantly, we saw striking differences in complement-dependent responses to HIT ULICs formed in the presence and absence of heparin. With heparin, HIT Abs bind to circulating PF4/H complexes forming soluble ULICs that bind to PMNs/monocytes in a complement-dependent manner to promote leukocyte-driven cytokine release and monocyte TF expression. Without heparin, HIT Abs bind to cell-surface GAGs on endothelial cells (ECs) and platelets, rather than white cells, forming cell-bound ICs that drive complement-mediated platelet adhesion and EC procoagulant activity. Based on these observations, we will test the hypothesis that complement responses to HIT ULICs are impacted by heparin itself and involve distinct cellular targets, activation pathways and effector mechanisms though the following specific aims: 1) Cell injury by soluble HIT ULICs requires complement and Fcγ receptors. We show the importance of the proximal complement pathway (C1/C3) for binding of soluble HIT ULICs to PMNs and monocytes via complement receptors (CRs). We hypothesize that binding of soluble HIT ULICs to CR1/CR3 initiates cell signaling and cytoskeletal reorganization, FcγR clustering, membrane scramblase and monocyte TF expression. 2) Cell-bound HIT ICs promote complement-dependent platelet adhesion and direct endothelial injury. We show terminal complement pathway mediated platelet adhesion to injured EC and upregulation of EC procoagulant activity by cell-bound ICs. We hypothesize that cell-bound HIT Abs initiate sublytic C5aR-mediated membrane injury leading to further upregulation of C5aRs, release of high molecular weight von Willebrand Factor multimers, and C5aR-dependent TF expression. 3) Complement activation as a biomarker for pathogenic HIT antibodies. Our studies show a strong correlation of complement activation with platelet-activating effects of HIT ULICs. We will test the hypothesis that complement activation identifies a subset of pathogenic HIT Abs by characterizing the biologic properties of HIT IgG Abs that do/do not activate complement and form soluble and cell-bound ULICs, develop complement activation as a biomarker assay, and examine complement’s role in the therapeutic efficacy of intravenous immunoglobulin. Together, these proposed studies of complement in HIT will advance our understanding of its underlying pathology, should improve diagnostic testing and position complement inhibitors as adjunctive therapy for this prothrombotic disorder.

NSF Awards · FY 2025 · 2025-01

With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Ivan A. Moreno-Hernandez of Duke University is studying strategies to improve electrocatalysts during the electrochemical production of valuable chemicals. The project will employ the synthesis and manipulation of metal oxide nanocrystals to explore chemical effects on catalysis and to improve catalytic properties. State-of-the-art methods that enable the observation of catalytic particles in liquid environments with transmission electron microscopy will be utilized to understand the formation of reactive species on individual nanocrystals. The research will advance next-generation catalysts and transform society by enabling sustainable chemical infrastructures using clean energy sources. The project integrates research in catalysis, electrochemistry, electron microscopy, and data science approaches to train students at all levels with a multidisciplinary skillset. The project will integrate research and education for English- and Spanish-speaking students through curricula that emphasizes electrocatalysis to train the next generation of STEM workers in sustainable chemistry. With the support of the Chemical Catalysis program in the Division of Chemistry, Professor Ivan A. Moreno-Hernandez of Duke University is studying strategies to improve iridium oxide-based nanocrystal electrocatalysts for the oxygen evolution reaction. Understanding the catalytic properties of nanocrystals remains a challenge due to the wide property dispersion among nanocrystals, and the lack of techniques to probe individual catalyst particle structures during catalysis. A central hypothesis that this project will explore is that compositional and dynamic disorder within otherwise pristine electrocatalysts could unlock new catalytic capabilities. The project will explore the influence of nanocrystal chemistry and disorder on catalytic performance, reaction intermediate binding energies, and active site restructuring via techniques that bridge spatiotemporal length scales. Single electrocatalyst particles will be observed at atomic resolution during catalysis with liquid phase transmission electron microscopy. The project can transform the frontiers of catalysis science by advancing strategies to design active electrocatalysts and obtaining knowledge of active sites that unify experiment and theory via a common spatiotemporal scale. 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.

NIH Research Projects · FY 2026 · 2024-12

Seasonable inlfuenza A and B result in up to 500,000 deaths globally each year, and influenza pandemics resulting form antigenic shifts have the potential to result in in orders of magnitude higher mortality. However, seasonable influenza vaccines are only 10-60% effective, and are unlikely to provide any protection against novel pandemic strains. To address these challenges, our long-term vision is to develop computational models that predict the sequences of antibody archetypes generated in response to an influenza protein sequence, and then predict via simulation ensembles the protection afforded by a polyclonal response — with obvious and direct application to mRNA and other vaccines. Our short-term goal for this proposal is to construct generative models of the antibody repertoire and mechanistic models to investigate how a polyclonal antibody response protects against infection. Here, our overarching hypothesis is that eliciting a diverse but coordinated humoral immune response will result in more effective and robust vaccines. To understand how a pluralistic humoral immune response, with diverse neutralizing and effector antibody mixtures, is generated and coordinated to protect against the influenza virus, computational modeling is essential. Hence, we propose to develop multiscale computational models calibrated on extensive infection and vaccination datasets. These computational models will guide vaccine development by providing the tools to generate anti-influenza antibody sequence candidates, identify effective defense mechanisms at the airway mucosa, and predict vaccination outcomes in heterogenous populations. In Aim 1, we propose to develop protein language models (PLM) to capture antibody repertoire information from BCR-seq and Ig-seq data sets as high-dimensional vectors in an embedding space, using samples from healthy, infected, and vaccinated human cohorts. By comparative analysis of the distribution of antibodies in the embedding space, we will identify sequence archetypes of the anti-influenza response, and experimentally characterize the biophysical and functional properties of synthetic archetypal antibodies. In Aim 2, we propose to develop agent-based models (ABM) to capture the spatiotemporal CODEX data sets of parenchymal cells, immune cells, antibodies, and virus from ex vivo experiments in lung explants from human donors. These models will enable in silico simulations to evaluate hypotheses for how innate and humoral mechanisms provide protection at the mucosal interface. In Aim 3, we propose to develop dynamical systems models to capture nonlinear interactions between the effector functions of a polyclonal antibody response and the impact on peak viral load and clearance time from qualified clinical immunoassays from human influenza challenge cohort data sets. The overall impact will be delivery of three complimentary antibody, agent-based, and compartmental computational platforms that together will rapidly accelerate the genesis of broadly effective seasonal and universal influenza vaccines.

NSF Awards · FY 2024 · 2024-12

The 28th Annual Quantum Information Processing Conference (QIP 2025), the premier international conference on the theory of quantum information science, is co-hosted by Duke University and NC State University in Raleigh, North Carolina, from February 24–28, 2025. This conference serves as a global platform for researchers in quantum computing, cryptography, quantum foundations, information theory, and many-body physics to present and discuss groundbreaking advancements. Despite the critical role of mathematics in quantum information science, U.S.-based mathematicians are currently less represented in the QIP community, hindering opportunities for broader interdisciplinary collaboration. To address this, the project provides 40 fellowships for U.S.-based students and early-career researchers in the mathematical sciences. These fellowships align directly with the NSF mission by fostering the progress of quantum information science, strengthening the workforce, and driving innovation in a field with broad societal and technological implications. Quantum information science has deep ties to mathematics, with foundational contributions from applied and computational mathematics, algebra, analysis, and probability among others. This project supports the participation of U.S.-based students and early-career researchers in the mathematical sciences by providing 40 travel fellowships to attend QIP 2025. The fellowships aim to empower talented young mathematicians to engage in cutting-edge research and interdisciplinary interaction opportunities with quantum information science researchers from other disciplines. Fellows will benefit from research talks, poster sessions, rump sessions, industrial sessions, networking opportunities, and interdisciplinary exposure, significantly advancing their professional development. Ultimately, this initiative supports the development of the next generation of quantum researchers, advancing the field and contributing to the NSF mission. 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.

NIH Research Projects · FY 2026 · 2024-12

PROJECT SUMMARY/ABSTRACT Invasive aspergillosis (IA) is a severe and frequent complication in immunocompromised patients. Earlier diagnosis expedites appropriate antifungal therapy and drastically improves clinical outcomes and patient survival. However, current pathogen-based fungal testing is often hampered by poor accuracy, delayed results, and a requirement for invasive sampling. This proposal will overcome these limitations by establishing novel, rapid, noninvasive diagnostic and prognostic biomarkers for IA based on host transcriptional responses to infection. Moreover, the project will ensure that these biomarkers can accurately function in the presence of diverse immunosuppressive regimens. Building on the candidate’s published work, the main goals of this project are three-fold: 1.) To derive and optimize a diagnostic host gene expression signature of IA from banked samples of a large established repository; 2.) To enroll and serially sample subjects with fungal infection to derive a prognostic host gene expression signature of IA and measure response to treatment; and 3.) To develop the candidate into a uniquely qualified independent translational physician-scientist, advancing non-invasive diagnostic and precision medicine tools for infection in immunocompromised patients. This research is critical for NIAID’s mission to better understand and treat fungal infections, particularly as Aspergillus species have become increasingly recognized as critical fungal pathogens. Additionally, the project will develop the candidate’s expertise in biological pathway analysis and fungal immunology, diagnostic test development and evaluation, and biomarker implementation. The results of this project will also generate preliminary data for the next step in this program of research, progressing through the translational pathway to develop and implement a clinically validated host-based diagnostic and prognostic tool for IA. The candidate’s research and career development will be guided by an invested and experienced mentorship team at a research-intensive institution with abundant resources in infectious diseases, translational research, transplantation, genomics, and precision medicine.

NIH Research Projects · FY 2026 · 2024-12

PROJECT SUMMARY Proper regulation of chromatin is essential for copying, maintaining, and transcribing DNA and errors in this machinery are frequently associated with a variety of diseases. A critical part of chromatin regulation are interactions between protein complexes and nucleosomes containing core histone subunits. Histones can be post-translationally modified and patterns of modifications, including acetylation, serve as a code to recruit and localize chromatin regulators through “reader” domains. Histone reader function and binding selectivity are incompletely understood and, with few exceptions, reader mutations observed in disease are not well characterized, despite being desirable therapeutic targets. To address these gaps in knowledge, we propose the development of an ultra-high-throughput platform called phage- and robotics-assisted near continuous selection (PRANCS), for structure-function analysis of all known readers, their mutations, and the histone modification interactome. We will use PRANCS to systematically characterize both the natural specificity and impact of mutations on reader domain interactions with acetyl-modified histone subunits. First, we will build improved robotics methods to parallelize PRANCS and optimize the scale of an individual assay (Aim 1). We will develop and apply a two-hybrid assay to study reader domain recognition of histone subunits modified using non-canonical amino acids (Aim 2). Finally, we will build and optimize the use of next-gen sequencing methods for enhanced quantitative readouts in PRANCS, applying this method to systematically assess the impact of thousands of reader mutations found in disease on histone modification binding (Aim 3). Results from this proposal will elucidate entirely new insights into the fundamentals of chromatin recognition and the functional impact of reader domain mutations, collectively building towards refined approaches in disease treatments.

NIH Research Projects · FY 2026 · 2024-12

ABSTRACT Chlamydia trachomatis (Ct) represents the most common sexually transmitted bacterium in the U.S., and is responsible for severe sequelae such as pelvic inflammatory disease, ectopic pregnancy, and infertility. No vaccine is currently available. Much of what makes Ct such an effective pathogen is its ability to evade the host immune response. The molecular mechanisms of this immune evasion remain largely uncharacterized, and represent a major gap in knowledge which is necessary for the proper prevention of this disease. Chlamydia species invade host epithelial cells, where they establish infection within a membrane-bound vacuole called an “inclusion”. At least two closely-related species of Chlamydia can infect human epithelial cells: C. trachomatis (Ct) and C. muridarum (Cm). Once inside these epithelial cells, Cm is vulnerable to clearance by intracellular defenses galvanized by interferon-γ (IFNγ), while Ct is able to expertly evade these defenses and will grow uninhibited. These IFNγ-induced defenses, and their evasion by Ct, are poorly understood phenomena in need of further research. We recently discovered that a human IFNγ-stimulated gene (ISG) drives restriction of Cm through an intracellular, proteasome-dependent mechanism. I also found that the Ct gene IncS confers protection against this ISG. In the following proposal, I propose experiments to better understand how Ct IncS drives evasion of human IFNγ-stimulated host defense. These experiments will include comparisons between Cm and Ct IncS function, and will focus on the role of two host proteins recruited by Ct IncS: STIM1/2. Our hypothesis is that Ct recruits STIM1/2 as a form of “molecular camouflage” to hide individual inclusions from host recognition and immune clearance. I will also propose experiments to better understand how the proteasome defends against Cm. I will perform several experiments to determine whether the proteasome directly localizes to Cm and if it degrades bacterial proteins to clear Cm infection. I will also perform a proteomics screen to identify additional protein partners involved in IFNγ-mediated restriction of Cm. Together, these experiments will provide a better understanding of intracellular defenses within the human epithelium and will characterize a novel mechanism of immune evasion by C. trachomatis. Through characterizing host-pathogen interactions during Chlamydia infection, we hope to guide the design of improved options to prevent and treat chlamydial disease.

NIH Research Projects · FY 2026 · 2024-12

ABSTRACT Staphylococcus aureus bacteremia (SAB) is a common and frequently lethal infection. Despite appropriate antibiotic therapy, many patients will experience treatment failure in the form of relapsed SAB (>1 episode with the same isolate) and persistent SAB (>5 days of persistently positive blood cultures). Antibiotic treatment failure is driven by the interplay of host, treatment and microbial factors. Despite decades of research focusing on treatment and patient factors, mortality from SAB remains at 25% and we still know very little about the role of microbial biology in treatment failure. Innovative approaches are required to address this important knowledge gap. To address this problem, Dr. Parsons’ long-term goal is to use patient clinical data, bacterial clinical isolates and focused genomic and laboratory studies to determine the link between bacterial biology on the bench and clinical outcomes in patients. The short-term goal of this proposal is to identify genomic polymorphisms influencing antibiotic treatment failure in SAB patients. The central hypothesis posits that specific genomic changes lead to an enhanced ability to evade the innate immune system and antibiotics, which drives worse clinical outcomes in patients. This will be tested through two specific aims. Aim 1 will identify bacterial genotypes associated with SAB antibiotic treatment failure using murine in-vivo evolution model. Dr. Parsons has developed several S. aureus strains with enhanced virulence and increased antibiotic tolerance via serial passaging through a murine bloodstream. Dr. Parsons will identify genomic changes within these isolates and examine the clinical relevance by screening for these mutations in a prospectively enrolled cohort of patients with relapsed SAB. Aim 2 will identify new bacterial genotypes associated with SAB antibiotic treatment failure. This aim will use a unique high-throughput sequencing and a bacterial Genome Wide Association Study (GWAS) pipeline combined with the world’s largest SAB biorepository to identify genomic variants associated with persistent SAB and relapsed SAB. The work will also focus on unraveling the biological mechanism driving persistent bacteremia of two polymorphisms previously identified using this GWAS technology; plsX and metN2. Identification of genes associated with antibiotic treatment failure in humans will increase our understanding of the devastating outcomes associated with SAB and identify new drug candidates aimed at reducing S. aureus virulence and sensitizing the bacteria to antibiotics. Dr. Parsons’ training to date has focused on the biochemistry of S. aureus metabolism. He will supplement this experience with training in genomics and observational clinical studies to focus on the link between bacterial genomic variations and clinical outcomes. The skills and mentoring acquired during this award will facilitate Dr. Parsons’ development into an independent physician-scientist focused on bacterial determinants of clinical outcomes in bacterial bloodstream infections.

NSF Awards · FY 2024 · 2024-12

This award will support two separate one-day virtual conferences entitled “Building Teams to Build Better Epidemiological Models: Balancing Participation from Mathematical and Social, Behavioral, and Economic Sciences” (https://sites.duke.edu/betterepidemiologicalmodelsconference/), to be held in January 2025. In a crisis such as the COVID-19 pandemic, mathematical models played their role in designing, developing, deploying, and evaluating public health strategies with different levels of success. Still, all were confronted with prioritizing public health or economic viability. To frame a sound pandemic response strategy, mathematical models are primary tools that must incorporate behavioral components and frameworks to be more efficient and useful for public health policy interventions and the evaluation of the economic impact of such measures. The COVID-19 pandemic highlights the need to develop mathematical methodologies, new techniques, and innovative approaches designed to incorporate the new paradigm of behavioral dynamics into the transmission dynamics of human diseases. Multidisciplinary teams are needed to innovate new mathematical methodologies which incorporate human behavioral and social dynamics. This award will be used to support a conference to bring together mathematical and social / behavioral / economic scientists to develop improved epidemiological models which can protect both public health and the economy. Applicants will be selected to balance these research areas, with attention given during the selection process to ensure that women and members of underrepresented groups are fully considered with an eye to broadening participation. The standard framework for the mathematical modeling of infectious diseases is the basic Kermack-McKendrick model, a compartmental model framed in ordinary differential equations and their extensions to stochastic and hybrid models. Mixing is a random process in this framework, and this characteristic has pervaded in models for prediction and forecasting and is one, but not unique, of the most challenging and important topics in modeling infectious diseases: how to modify the basic assumption of the homogeneous population in the model to incorporate significant behavioral effects robustly and effectively. For example, there have been several efforts in literature to integrate behavior; one of them is the one that assumes that agents that interact during the transmission of the disease are rational, i.e., the individuals behave in a way consistent with a rational evaluation of risks. This model type is based on economic thinking in which costs and benefits are balanced, where there is a trade-off that rational agents resolve. The problem in epidemiology is that many of the actions of natural agents during an epidemic do not adapt to this hypothesis; therefore, applying this type of modeling requires the development of innovative ideas, alternative conceptual frameworks, and new mathematical techniques and methodologies. Scientific teams which can innovate and parameterize mathematical models which are tractable, represent an analogue of human behavior and transmission, work across a variety of domains and settings, and can be used to test interventions are needed. 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.

NIH Research Projects · FY 2026 · 2024-11

The upper respiratory tract is lined by two distinct mucosal tissues. Respiratory mucosa humidifies and modulates incoming air temperature as it passes toward the lungs. Olfactory mucosa on the other hand is specialized for chemosensation and supports detection of airborne odorants by olfactory sensory neurons (OSNs). We have recently described an endothelial level barrier, a “blood-olfactory barrier” (BOB), that prevents high molecular weight proteins like antibodies (~150kD) from accessing the olfactory mucosa. Antibodies are absolutely required to prevent olfactory reinfection by airway viral infections, but serum neutralizing antibody provides no protection. Our studies have found that antibody secreting plasma cells (PCs) must reside within olfactory tissues- beyond the BOB- to provide locally neutralizing antibody in order to prevent olfactory infection. While vaccines generate robust circulating antibody titers, we found that they generally fail to differentiate OlfPC and as such, fail to protect against olfactotropic pathogens (that infect the olfactory mucosa), including SARS-Cov-2. Tissue specific PCs are generally understudied and this proposal will specifically establish a foundational understanding of when OlfPCs establish tissue residence, how long and where they reside within olfactory tissues, and how these cells relate to the peripheral BM PC pool. These data will empower mechanistic improvements to vaccines against upper respiratory infection.

NSF Awards · FY 2024 · 2024-11

NCShare AI-GaaS addresses two extremes on the demand spectrum for GPUs: first, GPUs are often unavailable or unaffordable to institutions with modest needs and budgets; and second, Artificial Intelligence research, including Large Language Model research, often requires GPUs at a scale even the largest research universities struggle to attain. By moving GPU sharing from the individual campus level to the state level the result is more efficient GPU utilization, greater aggregate computational capacity for solving research problems, improved economies of scale, and democratized access for smaller schools and under-resourced minority-serving institutions. NCShare AI-GaaS creates a powerful cluster of InfiniBand-interconnected Nvidia GPUs. The horizontally scalable cluster can be virtualized using the vendor's Multi-Instance GPU (MIG) with the slurm scheduler or their vGPU service and Kubernetes. These enable allocation and dynamically-configured access at the sub- or multi-GPU level, in support of a wide range of research and education needs. Operating atop North Carolina's Research and Education Network, the cluster leverages two prior NSF awards that created a state-wide Shared Science DMZ and a shared High Performance Compute cluster providing virtualized software stacks. Combining existing NCShare federated access control with Confidential Computing capabilities on the GPUs, NCShare AI-GaaS can deliver secure, authorized GPU access to colleges and universities throughout North Carolina. While the project is designed to serve general (non-regulated) research projects, a reference implementation using encrypted overlay networks and SDN techniques may enable use for certain regulated research projects. 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.