Johns Hopkins University

NSF Awards · FY 2026 · 2026-08

Many scientific breakthroughs happen when a researcher notices that a technique from one field could solve a problem in another — a discovery in materials science, for example, might share a deep structural similarity with a finding in a different field. Yet with hundreds of new papers appearing each month even within narrow specialties, no individual can read broadly enough to spot these cross-cutting connections reliably. This project develops artificial intelligence systems that help researchers uncover such hidden connections across the scientific literature — systems that can recognize structural parallels between ideas in different domains. Critically, creative reasoning by AI carries risks: a connection that looks insightful may turn out to be unfounded. To address this, the systems developed in this project are designed to be transparent, showing researchers the evidence behind every suggestion so they can verify claims and distinguish well-supported insights from speculation. This project serves the national interest by promoting the progress of science. By helping researchers identify promising new directions more quickly this work has the potential to accelerate discoveries that directly improve quality of life. All tools and datasets will be released publicly, and in partnership with major research conferences, these tools will be integrated into the peer-review process to support real-world scientific evaluation. The project also includes substantial educational efforts: paid summer research internships for Baltimore high school students, a new AI literacy course for pre-college students, integration of research tools into undergraduate teaching, and inclusive workshops and mentoring to broaden participation in computing and artificial intelligence. This project develops interpretable, large language model (LLM)-based frameworks that support creative scientific discovery while ensuring transparency and grounding in evidence. The research is organized around four interconnected thrusts. The first thrust develops a general-purpose formalism for analogical retrieval over large scientific corpora. This uses an iterative approach to create a retrieval-and-reasoning loop, akin to exploration-exploitation, that enables applications including cross-domain hypothesis generation and tabular summarization of related literature. The second thrust advances temporal reasoning in scientific domains. It begins by systematically evaluating how reliably LLMs encode scientific knowledge over time, analyzing how factors such as publication age, citation impact, and controversy influence what models have memorized. Building on these findings, the thrust develops methods to mine the scientific genealogy of research papers, tracing chains of intellectual influence through recursive analogical retrieval to produce interpretable graphs of how ideas emerge, evolve, and connect. The third thrust creates a principled framework for grounded, verifiable generation. The core innovation is a "myopic" verifier, a secondary model that simulates a time-constrained human fact-checker, assessing each claim using only a narrow excerpt from the cited source. Models are trained via reinforcement learning to produce only claims that are immediately and explicitly verifiable. This approach extends to incorporating verbatim quotations from cited evidence, further reducing the cognitive burden of verification. The fourth thrust builds publicly available scientific tools and evaluation benchmarks, including retrospective analogy benchmarks, grounding evaluation protocols, and a peer-review co-pilot to be deployed at major research conferences. The resulting methods are developed in collaboration with domain experts at the National Institutes of Health, but are broadly applicable to any domain requiring structured, cross-domain reasoning. 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 2026 · 2026-07

In recent years, the world has witnessed significant progress in optimization for emerging fields, including meta-learning, fine-tuning, automated hyperparameter selection, continual learning, fair batch selection, adversarial learning, and artificial intelligence (AI)-aware communication networks. Problems arising from these fields often exhibit a common nested optimization structure, which has motivated the study of bilevel optimization. However, there are many theoretical and computational challenges in large-scale bilevel optimization problems, e.g., those arising from machine learning on massive amounts of data in high-dimensional feature domains that have manifold constraints. This project will provide a comprehensive study of bilevel optimization theory, algorithms, and applications for large-scale problems. The outcomes of this project will benefit researchers in academia, government labs, and industry aiming to solve large-scale nested optimization problems in science and engineering. New applications in information science, signal processing, communications, statistics, and machine learning will be studied. This project consists of three intertwined thrusts. The first thrust focuses on developing fast and scalable Hessian-free bilevel algorithms with convergence rate guarantees. Specifically, several Hessian-free approaches will be designed and analyzed using methods of fully single-loop momentum, finite-difference matrix-vector estimation, and residual response Jacobian estimation. The second thrust aims to develop primal-dual, primal, and pessimistic bilevel methods, in addition to the analysis of convergence in the difficult case where no unique lower-level solution exists. In the third thrust, the investigators will develop algorithms for solving bilevel problems on non-linear manifolds and analyze the associated convergence of these algorithms. The developed algorithms will be implemented in the context of real-world applications, including fairness-aware machine learning, continual learning, resource allocation over communication networks, hyperparameter selection of principal component analysis, and dictionary learning models. 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 2026 · 2026-07

Conversion of low-value waste into valuable products using microbes can help strengthen the domestic supply of important chemicals. One such product is caproate, a medium-chain carboxylic acid used in fuels, animal feed, cosmetics, and other applications. However, making caproate from waste remains difficult. Challenges include low yields, poor selectivity, and unwanted byproducts. Synthetic microbial consortia, which are groups of microbes selected for specific waste types and end products, offer a promising solution. However, effective strategies are needed to make these systems stable and efficient. This CAREER project will develop engineering and modeling tools to design microbial consortia that efficiently convert dairy and brewery waste into caproate. The project will also train students at multiple academic levels, increase public awareness about waste upcycling, and strengthen partnerships among universities, industry, and local communities. Harnessing microbial consortia for waste valorization presents a considerable opportunity to shift from traditional waste treatment towards resource recovery. Monocultures often lack the enzymatic breadth for multi-step waste conversion, while mixed cultures suffer from poor selectivity and byproduct formation. Synthetic microbial consortia offer a modular alternative to conventional mixed culture or pure culture for biomanufacturing. Despite their high potential, the application of such synthetic microbial consortia is limited by challenges in selecting optimal microbial partners that best metabolizes complex waste from a wide pool of microbes, predicting interspecies metabolic interactions that shape consortia’s function, and maintaining long-term functionality and stability in non-sterile waste. To address these challenges, this CAREER project will engineer high-yielding synthetic microbial consortia for biomanufacturing of caproate from waste through three integrated research objectives: (1) develop a high-throughput, community-scale metabolic modeling framework to simulate interspecies interactions and guide selection of highest caproate producing synthetic microbial consortia for experimental validation; (2) engineer hydrogel encapsulation to enhance retention, stability, and functionality of synthetic consortia by fine-tuning composition and density of the polymer and cross-linker; and (3) evaluate the productivity and competitive advantage of encapsulated consortia for bioconversion of organic waste streams under environmentally relevant conditions. These research efforts will be closely integrated with a comprehensive education plan that includes 1) project-based workshop on engineering design and programming for high school students; 2) creation of an educational video highlighting student-led research and partnerships with local industries to foster public engagement and scientific literacy in waste upcycling; and 3) integration of innovative modules into undergraduate and graduate curricula. Together, these efforts will advance environmental biotechnology and develop skilled workforce to enable scalable and resilient biomanufacturing from waste. 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 2026 · 2026-07

Artificial intelligence systems rely on large groups of computers working together, but communication between these machines is becoming a major bottleneck. Today, many systems send the same data repeatedly between machines, which wastes time, energy, and network capacity. This project develops new ways for computers to share information more efficiently by using multicast, a method that allows one machine to send data to many others at once, much like a radio broadcast. The goal is to make communication faster, more reliable, and easier to manage in large-scale computing systems that support modern artificial intelligence applications. This project addresses the scalability and observability challenges that currently prevent the deployment of multicast in production environments. The technical work is organized into three integrated thrusts. The first thrust develops a scalable data plane using topology-aware algorithms to construct efficient transmission trees and introduces a new way to compress network forwarding state to fit within the limited memory of standard hardware. The second thrust creates an introspectable control plane and monitoring system that uses advanced probes and machine learning to detect and localize hidden network failures in real time. The third thrust integrates these research findings into the university curriculum through the creation of hands-on laboratory modules and a structured mentoring pipeline for students. This project improves the efficiency and reliability of artificial intelligence infrastructure that supports applications such as healthcare, science, and large-scale data analysis. By reducing communication overhead, it lowers energy consumption and operational costs in datacenters. The project also contributes to workforce development by integrating research into hands-on educational modules, mentoring students, and providing practical experience with real-world systems. Open-source tools, datasets, and experimental platforms produced by the project will enable broader adoption in both academia and industry, strengthening the overall computing ecosystem. 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 · 2026-06

SUMMARY We propose a three-day workshop, “Open-source Low-field MRI: Learn, Build, Scan,” to train biomedical imaging scientists through hands-on construction, scanning, and AI-based image reconstruction using low-cost, open- source MRI systems. Building on the successful “DELTA DIY MRI” at Johns Hopkins, it supports NIBIB’s leadership in biomedical imaging and NINDS’s neurotechnology workforce goals. The workshop’s three phases—Learn, Design & Build, and Scan—combine lectures, hardware prototyping, and real-time scanning with modular MRI setups. Daily milestones include RF Spin Echo and 1D projection (Day 1), 2D Turbo Spin Echo with deep learning super-resolution (Day 2), and 3D imaging with EMI mitigation (Day 3). Mentors with publication records will teach subsystems from magnets to phantoms. Luminary speakers will provide historical context and updates, fostering an engaging environment. Introductory sessions in GitHub and 3D printing will help participants of varied experience levels. Participants will use and contribute to open tools like PyPulseq, Virtual Scanner, and MRI4All, promoting reproducibility. The workshop plans to host 50 participants, with early registration, accessible venues, safety protocols, and NIH-compliant conduct policies. Attendees will gain access to detailed schematics, code, hardware docs, and recordings. The program expands access to MRI by enabling resource-sharing and providing post-workshop materials. Post-event support includes a resource hub, virtual mentoring, and remote scanner access. Promotion efforts include social media, society postings, and networks, with dissemination via project kits, walkthroughs, and tutorials. Scientific outputs include shared builds, software, and data reuse. Strategic backup plans address disruptions. Organizers—with experience and contributions to open-source MRI tools like PyPulseq—lead this initiative. The workshop aligns with PA-25-080 goals: technology dissemination, training, and research strengthening. It supports NIBIB’s imaging, workforce, and AI priorities, and NINDS’s neuroimaging goals. Deep learning super-resolution and phantom validation advance imaging rigor. Contextual understanding enhances critical thinking. Participants are encouraged to share knowledge, expanding capacity. Modular design allows replication in underserved regions. This scalable model aims to democratize MRI science and foster sustainable community growth, empowering technically fluent scientists to lead the future.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Antiretroviral therapy (ART) effectively suppresses HIV replication but fails to eliminate the viral reservoir—a heterogeneous population of infected cells that persist despite treatment. While most proviruses are defective, many retain the ability to express RNA, proteins, and even viral particles, contributing to persistent viremia, chronic immune activation, and complicating the interpretation of viral load measurements. Among these, proviruses with defects in the 5’ leader (5’L) region are common, and they decay more slowly than intact proviruses, are translation-competent, and are often implicated in persistent non-suppressible viremia (NSV), making them a critical component of HIV persistence and a significant source of residual antigenic stimulation. The long-term goal of this project is to define the mechanisms by which 5’L-defective proviruses evade negative selection, characterize their contribution to persistent viremia, and determine their impact on HIV-specific immune responses. Our central hypothesis is that 5’L-defective proviruses retain the ability to express viral genes while evading immune-mediated killing and cytopathic effects. To test this, we will (1) apply a novel digital PCR assay that distinguishes intact from 5’L-defective RNA without requiring sequencing to quantify their contribution of defective proviruses to plasma HIV RNA; (2) isolate and phenotype clonally expanded CD4+ T cells harboring 5’L-defective proviruses using two complementary approaches—stimulation with Tat-LNP followed by p24+ cell sorting, and limiting dilution culture ex vivo—to assess viral transcription and translation through multi-omic analyses (SMARTseq, Ribo-seq); (3) investigate susceptibility of 5’L defective proviruses to HIV-specific CTL, and leverage longitudinal samples from individuals with NSV to evaluate how antigen expression from defective proviruses shapes HIV-specific immune responses and clonotype dynamics over time. This research is innovative in developing a first-in-class assay to parse 5’L-defective RNA without sequencing and applying cutting-edge single-cell and immunologic profiling to study the interplay between defective proviral expression and host immunity. The proposed research is significant because it is expected to address critical gaps in our understanding of HIV persistence, improve the interpretation of HIV cure clinical trials, and inform new strategies aimed at targeting cells with intact and defective proviruses. Ultimately, this project has the potential to offer new tools to improve the clinical outcome of people living with HIV and opportunities for the development of innovative strategies to achieve HIV remission.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Prostate cancer is the second leading cause of cancer-related death in men and remains largely resistant to immunotherapy due to its immunologically cold tumor microenvironment (TME). However, benign prostate lesions, such as proliferative inflammatory atrophy (PIA), are characterized by active immune infiltration. Prior work has established that MYC acts as a molecular switch, initially promoting immune activation in early lesions before driving immune suppression in invasive disease. While MYC’s oncogenic role is well known, the transcriptional and chromatin remodeling mechanisms underlying this immune switch remain poorly understood. We hypothesize that MYC-driven reprogramming is mediated through SMARCD2, a critical subunit of the BAF chromatin remodeling complex, orchestrating differential binding of key transcription factors (AR, ERα, ERβ, MYC) to modulate the immune response within the prostate TME. Our long-term objectives are to understand the molecular dynamics underpinning prostate cancer progression from immune activation to immune escape and leverage this knowledge to inform the development of novel therapeutic strategies aimed at reversing immunosuppression and increasing available treatment options for advanced prostate cancer. To achieve this, Aim 1 will define transcription factors driving immunogenic vs. Immunosuppressive programs in prostate cancer by comparing benign proliferative atrophy (PIA) lesions to invasive carcinoma. We propose a model where early MYC activation fosters immune activation, whereas sustained MYC activation drives immunosuppression. Using coupled single nucleus ATAC- and RNA- seq on human prostatectomy samples, we will profile transcriptomic and chromatin landscapes to identify candidate regulators and networks associated with distinct immune states. Furthermore, Aim 2 will determine the tumor cell-intrinsic mechanisms by which MYC-driven chromatin remodeling initiated immunosuppression, focusing on the BAF complex subunit SMARCD2’s role in reprogramming both androgen and estrogen receptor activity. Hi-Plex CUT&Tag assays will reveal genome-wide binding patterns of AR, ERα, ERβ, and MYC in later stage prostate cancer cells. Additionally, functional studies using CRISPR/Cas9 mediated knockout and lentiviral-inducible overexpression of SMARCD2 followed by bulk RNA-seq and CUT&Tag assays in prostate cancer cell lines, will investigate SMARCD2’s regulatory role in altered transcription factor occupancy and immunosuppression. Collectively, this proposal integrates cutting-edge genomic techniques with robust molecular characterization to define novel transcriptional and epigenetic mechanisms underlying immune modulation in prostate cancer. This innovative approach will significantly advance our understanding of prostate cancer biology and identify novel therapeutic targets to improve clinical outcomes for prostate cancer patients.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Volumetric muscle loss (VML) is a critical sized skeletal muscle injury in which the natural healing mechanisms are overwhelmed, and loss of normal muscle structure and function persists over the lifetime of the patient. Successful regeneration of functional, integrated skeletal muscle tissue following VML requires the orchestration of satellite cell proliferation, myogenic differentiation, angiogenic and neuronal ingrowth, and the formation of neuromuscular junctions (NMJs). Numerous biomaterial- and cell-based strategies have been developed that exhibit significant potential to regenerate skeletal muscle but the restoration of functional neuromuscular junctions remains a major hurdle. After VML, the loss of muscle tissue disrupts this bi-directional nerve-muscle signaling that establishes and maintains NMJs in the remaining muscle. There is progressive denervation of NMJs, resulting in NMJ fragmentation and abnormal innervation. One caveat of current pre-clinical VML models is that they fail to recapitulate these progressive deficits inherent in chronic injury, which is typical of the clinical presentation. In published mouse studies, the VML injury is treated immediately following the surgical removal of tissue (acute injury). To phenocopy the clinical scenario we will establish a model of chronic VML with middle- aged (10 – 14 months old) mice that we will use to test our novel bioengineering therapeutic strategy. The goals of this study are to: (1) Use functional testing and advanced imaging techniques to characterize the neuromuscular deficits and establish a robust, validated chronic injury model; (2) Employ multi-tissue transcriptomic analyses to evaluate nerve-muscle cross-talk after VML to identify novel therapeutic targets for promoting neuromuscular regeneration and confirm the mechanistic basis of these targets using in vivo CRISPR knockouts; and (3) Develop advanced composite tissue engineered muscle grafts informed by imaging and transcriptomic data that promote muscle regeneration and restore functional neuromuscular junctions. We hypothesize that by identifying and targeting the molecular maladaptations inhibiting NMJ regeneration in chronic VML, we can establish more robust and effective biomaterial-based therapies. Overall, the study will provide a more stringent model of VML to be adopted by the field and integrate cutting edge imaging, transcriptomic, gene-editing, and nanotechnology tools to establish novel, effective, biomaterial therapies.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Tuberculosis (TB) is a global health emergency and is a major cause of morbidity and mortality worldwide. As a social determinant of health, stigma associated with TB affects those with the disease, their families and communities. TB stigma is prevalent, with global estimates ranging from 25-80% of people with TB experiencing TB stigma. It is associated with sub-optimal outcomes throughout the TB cascade of care, including reduced care-seeking, decreased treatment adherence and increased loss to care and mortality. The World Health Organization has identified combatting TB stigma as one of the top ten priorities necessary to meet the ambitious End TB targets—reducing global incidence by 80% and TB deaths by 90% by 2030. TB stigma also has important downstream effects on TB disclosure. Disclosure to close contacts who are at high risk for infection and active disease is a critical step in successful TB contact investigation. The WHO recommends conducting contact investigations among contacts of all newly identified individuals with TB, even in low- and middle-income settings. Contact investigation disrupts TB transmission by finding new cases earlier (decreasing the potential time of transmission) and by preventing new cases (reducing the reservoir of transmission). Although effective at detecting new cases of TB, implementation has substantial room for improvement—in two large trials of contact investigation conducted by our teams, participation of individuals with TB was only 60%. Qualitative interviews with individuals with TB, their close contacts and healthcare workers in the PI’s current K01 award in South Africa (K01HL151977) all highlight non-disclosure and its interplay with TB stigma as an important barrier to contact investigation. Lack of disclosure from an individual with TB means fewer contacts can be screened for TB, that some diagnoses among contacts will be delayed or missed entirely, and that those with TB may receive less support and will thus be likely to be retained in care. Despite the importance and interrelation of TB disclosure and associated stigma within the continuum of TB care, there are currently several profound knowledge gaps which inhibit our ability to design interventions to mitigate TB stigma and facilitate disclosure – 1) our understanding of TB stigma as a key determinant of health and factor affecting TB care is poor and lags behind those of other health conditions such as HIV in similar settings; 2) Few studies have examined the causes and downstream effects of non-disclosure among individuals with TB and their close contacts; 3) There is no standard instrument to measure disclosure; 4) There are no existing interventions designed to TB facilitate disclosure. In this proposal we seek to address these first three barriers through two separate systematic reviews of the literature- one on TB disclosure and one on TB stigma. This proposal expands on the PI’s K01 research goals to improve the effectiveness of TB contact investigation in high burden settings.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Any person that has broken a bone can attest to the exquisite pain that results. The central mediator of pain sensation is NGF (Nerve growth factor), which transmits nociceptive signals either by directly activating TrkA (Tropomyosin receptor kinase A) sensory neurons or through indirect mechanisms. An expanding body of literature suggests that TrkA+ sensory nerves have important efferent regulatory functions in skeletal repair. However, the precise molecular mediators of nerve-bone crosstalk during fracture repair have remained essentially unknown, and the possibility of therapeutic intervention remains unexplored. In order to address this knowledge gap, we recently validated new methods in retrograde peripheral nerve tracing and single cell RNA sequencing of sensory neurons that innervate bone. Our results have identified a unique transcriptomic profile of skeletal-innervating neurons, and uncovered their dynamic response to bone fracture. Multi-tissue scRNA-seq of skeletal innervating neurons and their target cells within the fracture callus elucidated potential mechanisms of neuronal regulation of fracture healing, including neuron-derived fibroblast growth factor 9 (FGF9). These observations have led to our central hypothesis that TrkA+ peripheral sensory nerves positively regulate the early response to traumatic bone injury by releasing FGF9 to stimulate skeletal cell proliferation and osteogenic differentiation, and that this interaction can be leveraged therapeutically to enhance fracture repair. Aim 1: Assess the role of sensory neuron derived FGF9 signaling in nerve-bone crosstalk during fracture repair. Here, we will perform experiments with peripheral neuron specific Fgf9 deletion (AdvillinFgf9) or AAV- mediated DRG neuron Fgf9 knockdown, followed by fracture phenotyping. In Aim 1B, rescue experiments will be performed in which local rFGF9 will be applied to the fracture site to reverse the effects of denervation. We hypothesize that sensory nerves stimulate periosteal cell proliferation and/or osteogenic differentiation in a FGF9 dependent process, ultimately positively regulating fracture repair and improving biomechanical integrity. Aim 2: Enhance fracture repair via promotion of sensory nerves using a locally delivered small molecule TrkA agonist. Here, we will optimize a percutaneous local injury site treatment using a novel hydrogel delivery system to enhance bone repair. A dual-release GA and BMP2 containing hydrogel/microparticle formulation is designed to improve early sensory neuron ingrowth via burst release, followed by enhancing osteogenesis via sustained BMP2 release. We hypothesize that the biomimetic GA local delivery system with or without BMP2 will enhance sensory neuron ingrowth and survival, and enhance fracture repair without induction of pain.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY/ABSTRACT Allergic inflammation of the airways leads to sneezing, coughing, excessive secretions, and, in some individuals, airway narrowing. Each of these effects can be mimicked by activation of sensory (afferent) nerves, particularly vagal (or trigeminal) nociceptive C-fibers terminating in the airway mucosa. Activation of these fibers results in both conscious sensations, such as the urge to sneeze or cough, and subconscious parasympathetic reflexes, including secretions and airway smooth muscle contractions. Allergic inflammation not only leads to overt C-fiber activation but also increases C-fiber “excitability”, such that response to a given inflammatory or mechanical stimulus are enhanced. When a C-fiber stimulus (e.g., an inflammatory mediator) activates its cognate receptor at the terminal, it causes a membrane depolarization known as the generator potential. The generator potential is physiologically relevant only if it is sufficiently large and fast enough to activate voltage-gated sodium channels (NaV1s), which are responsible for generating action potentials that are conducted to the central nervous system (CNS). Until recently, these channels could only be blocked by large concentrations of low-affinity, non-selective sodium channel blockers (i.e., local anesthetics), which must be administered locally for safety reasons. We now know that there are nine NaV1 subtypes (NaV1.1 – NaV1.9). Using single-neuron RT-PCR and RNA-seq in vagal ganglion neurons projecting fibers to mouse airway mucosa, we have found that airway C-fiber neurons almost exclusively express NaV1.7, Nav1.8, and NaV1.9 mRNA. These channels are largely restricted to C-fiber sensory neurons. Highly selective blockers for NaV1.7 and NaV1.8 are currently in development, and a NaV1.8 blocker has recently been approved for pain treatment. In the K99 portion of this proposal, I aim to understand these channels using genetic manipulations and highly selective pharmacological inhibitors to provide direct electrophysiological evidence that blocking each of these NaV1 subtypes can, via distinct mechanisms, inhibit action potential discharge and conduction in airway C-fibers from both healthy and allergen (cockroach extract)-induced inflamed airways. In the R00 portion of this grant, I will expand the scope of this study to in vivo reflex physiology experiments, specifically evaluating airway hyperreactivity in mice while selectively blocking NaV1.7, NaV1.8 and NaV1.9. The extent of inhibition suggests that blocking NaV1.7, NaV1.8, Nav1.9, or a strategic combination thereof may represent a novel and effective approach to treat allergy-associated coughing and sneezing, as well as reflex secretions, nasal congestion, and bronchoconstriction. The training received under this proposal will facilitate my transition to leading my own independent academic research laboratory focused on decoding neuromodulation mediated by NaVs in the inflamed airway.

NIH Research Projects · FY 2026 · 2026-06

Project summary/abstract: This application proposes to replace an end-of-life Philips research 3T MRI in the MRI Service Center at Johns Hopkins University (JHU) with a state-of-the-art Philips MR7700 3T MRI system. The proposed MRI scanner will benefit NIH-funded investigators across JHU and beyond. The current Philips research 3T MRI system has supported numerious NIH-funded projects over the past 20 years but is now reaching its end-of-life. The vendor, Philips Healthcare, has informed us that, due to the age of the hardware and the scarcity of this model among their customers, service contracts or replacement parts will no longer be available for this system beyond December 31, 2025. Therefore, the replacement with a new research 3T is both timely and imperative. JHU’s MRI technical group is one of the largest in the nation (>150 people). Therefore, considerable technical expertise is available among the Center personnel and other investigators at Johns Hopkins Radiology. Furthermore, JHU is known for its translational research and many clinicians have a strong need for reseach MRI as part of their NIH-funded studies. The long-term objectives of the proposed new research MRI are therefore to support NIH-funded projects at JHU across a wide range of disciplines, facilitating improved understanding of diseases and searching for a potential cure. For this application, we have assembled 12 Major Users and 5 Other Users consisting of different organs or clinical conditions (from 8 different NIH ICs). The proposed new research 3T is a Philips MR7700. This decision was made after a careful survey of similar systems by other vendors and a thorough assessment of sequence compatibility and continuity, especially the needs of our Major and Other Users. The new system has several important features when compared to the old system and other products on the market. The system is equipped with digital Stream (dStream) technology in which digital sampling occurs in the RF coil. This results in a SNR increase by 40% which will benefit all Major and Other Users’ projects. The new system has a 70cm bore which is less claustrophobic for patients, causes less discomfort, can accommodate larger patients, and allows a greater field-of-view. This is especially important for Major Users’ projects that involve vulnerable or high-BMI patients. The new MRI system uses a zero-boil off magnet and requires 0 refill of helium. Thus this system is environmentally more sustainable and reduces costs in maintainence and service contract. The new system is equipped with AI and deep learning technology are included in protocol suggestion, automatic planning, and image reconstruction. This improved efficiency in workflow will give all Major Users more time for actual scanning, allowing higher SNR or resolution. Simultaneous multi-slice (SMS) technology will allow faster fMRI and diffusion acquisitions, which are particularly beneficial for brain MRI studies. Collectively, these improved technologies and new features will significantly enhanced the projects of our Major and Other Users.

NSF Awards · FY 2026 · 2026-06

Data-driven personalized decision-making has become increasingly important across many fields, such as health sciences where tailoring treatments to individual patients can improve effectiveness and reduce adverse effects. Achieving reliable personalized decisions requires understanding cause-and-effect relationships between actions and outcomes. However, most real-world data sources, such as electronic medical records, health surveys, and social media data, are observational rather than randomized, making causal relationships difficult to establish. In these settings, hidden or unmeasured factors may influence both the actions individuals take and the outcomes they experience, leading to biased conclusions and unreliable recommendations if not properly addressed. This project will address this fundamental challenge by developing new statistical methods for learning optimal personalized decision rules from observational data when important confounding factors are not fully observed. The project will consider both single-stage and sequential decisions, with particular attention to continuous treatments such as medication dosages. A motivating application is kidney transplantation, where optimizing immunosuppressive therapy over time is essential to reduce the risk of graft failure while minimizing harmful side effects. By enabling more reliable individualized decision-making, this project will advance statistical science, machine learning, and artificial intelligence, support the training of students in modern data science, and contribute to improved health outcomes and broader societal well-being. This project aims to develop novel Bayesian causal methods for estimating treatment effects and optimizing individualized decision rules from observational data with unmeasured confounding. A Bayesian joint modeling framework will be introduced for treatment, outcome, observed covariates, and latent confounders, leveraging mild distributional assumptions to enable causal identification without relying on additional data sources required by many existing approaches, such as instrumental or proxy variables. The project will also develop a dynamic Bayesian causal modeling framework for longitudinal data, where treatment decisions and unmeasured confounders evolve over time. This framework will support the estimation of adaptive treatment regimes that respond to an individual’s evolving treatment history, outcomes, and characteristics. In addition, the project will design optimization methods for both single-stage and sequential decision-making, using posterior uncertainty to improve robustness in finite and unbalanced observational data settings. The methods will be evaluated through simulation studies and applied to large-scale real-world kidney transplantation data for studying optimal personalized and dynamic immunosuppressive dosing strategies. To facilitate broad dissemination, open-source software will be developed for implementation. The resulting framework and tools will provide a general approach to reliable personalized decision-making in biomedicine and other fields that rely on complex observational data. 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 · 2026-06

PROJECT SUMMARY: There is a need for a new immune engineering platform technology to engineer T cell responses and direct their function for diseases spanning from viral infection to cancer. Most vaccination strategies rely on delivering the antigen of interest along with a variety of adjuvants to enhance the response. We have developed nanoparticles which deliver costimulatory molecules and proinflammatory cytokines to tumors that can function in an antigen agnostic manner to generate potent local and systemic immune responses. However, technologies capable of directly boosting antigen presentation to enhance viral and antitumor immunity are lacking. We have developed a library of biodegradable poly(beta-amino ester) (PBAE) nanoparticles (NPs) capable of delivering a variety of nucleic acid cargoes to specific cell types and identified initial formulations to specifically deliver genes to antigen presenting cells (APCs) and tumor cells. Here, we will develop novel antigen- presentation modulating (APM) NPs, enhancing MHC expression, and engendering vaccination strategies for broad applications to cancer, infectious disease, and autoimmunity. To develop these APM-NPs, we will systematically test whether NP-mediated delivery of plasmid DNAs encoding MHC transactivators and/or interferon pathway genes can enhance antigen presentation in tumor cells (Aim 1a) and TLR agonists, immunoproteosome components, costimulatory molecules, and transcription factors that polarize dendritic cells can improve antigen presentation in APCs (Aim 1b). To develop even more improved formulations specific for APCs and tumor cells, we will synthesize a library of biodegradable polymers based on linear, branched, and lipophilic PBAEs and screen the library for efficient cell type specific delivery (Aim 1c) and confirm efficiency and specificity in vivo (Aim 1d). Using our already identified formulations, we will combinatorially evaluate whether tumor- or APC-directed delivery or a combination can enhance antigen specific T cell responses in co-culture models in vitro (Aim 2a) and thereby identify the most promising candidates to evaluate in vivo. Using an adoptive transfer model to allow us to track antigen specific T cells, we will then evaluate the impact of optimized APM-NPs to modulate local and systemic antitumor immune responses using high dimensional flow cytometry and a novel multiplex immunofluorescence panel optimized for murine FFPE tissues (Aims 2b, 2c). Finally, to test whether the optimized combinatorial APM-NP approach can synergize with immunotherapy and lead to a therapeutic benefit, we will evaluate whether local delivery of APM-NPs, in combination with checkpoint inhibition, can treat two mouse models of cancer (Aim 2d). As this technology is fundamentally flexible, if successful, the proposal will lay the foundational groundwork for a novel platform technology to genetically modulate antigen presentation and enhance vaccination strategies, with applications to viral vaccination, cancer immunotherapy, and beyond.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Iron-sulfur (Fe-S) clusters are essential protein cofactors for a wide range of cellular processes, including respiration, DNA repair, and metabolism. Deficiencies in Fe-S cluster biosynthesis underlie numerous human diseases, such as Friedreich's ataxia (FA), and contribute to mitochondrial dysfunction and cellular stress. Central to Fe-S cluster assembly is the transfer of a sulfane sulfur, typically in the form of a persulfide, from cysteine desulfurase enzymes (e.g., NFS1) to scaffold proteins (e.g., ISCU2), where cluster formation occurs. Recent structural and biochemical studies have elucidated the mechanism of persulfide transfer and highlighted the critical role of persulfide intermediates in cluster assembly and maturation. However, disruptions in this transfer pathway can impair Fe-S cluster formation, leading to cellular dysfunction. FA is a debilitating neurodegenerative disorder caused by a deficiency in frataxin (FXN), a protein that catalyzes a critical sulfane sulfur transfer for Fe-S cluster assembly. This FXN deficiency leads to impaired Fe-S cluster biosynthesis, mitochondrial dysfunction, and heightened oxidative stress. This proposal will investigate the protective and restorative effects of hydropersulfide (RSSH) donors in FA patient-derived skin fibroblasts and compare those effects in control cells. Aim 1 will investigate how RSSH donors rescue defective Fe-S cluster biosynthesis in FXN-deficient cells. We will analyze whether RSSH donors restore Fe-S cluster formation in FA patient-derived skin fibroblasts (which are FXN-deficient), examining their impact on FXN levels, mitochondrial aconitase activity, ISCU2 persulfidation, and ferredoxin levels, all critical to Fe-S cluster assembly. Aim 2 will elucidate the role of RSSH donors in improving mitochondrial biogenesis and function. Persulfidation has been linked to enhanced mitochondrial bioenergetics and biogenesis, with mitochondrial persulfides supporting electron transport chain activity and membrane potential. We will evaluate mitochondrial biogenesis and function focusing on PGC-1a and SIRT3 signaling, mitochondrial count, membrane potential, and respiratory function/ATP production following RSSH donor treatment. Aim 3 will assess whether RSSH donors restore intracellular RSSH levels and prevent iron-mediated oxidative damage/ferroptosis in FA cells. Low RSSH levels in FA cells leave them susceptible to oxidative damage and iron overload in FA contributes to this oxidative damage and triggers ferroptotic cell death, exacerbated by dysregulated ferritinophagy. We will assess whether RSSH donors mitigate these processes by restoring redox balance and iron homeostasis, thereby ultimately reducing cellular vulnerability to ferroptosis. Collectively, these studies will provide mechanistic insights into how RSSH donors combat defective Fe-S cluster biosynthesis and alleviate mitochondrial dysfunction, oxidative damage, and iron toxicity.

NIH Research Projects · FY 2026 · 2026-06

Treatment with immune checkpoints inhibitors has had unprecedented responses in turning some deadly cancers into chronic diseases. Despite these successes, 80% of cancers including prostate cancer (PCa) have complex tumor microenvironments (TMEs) that are resistant to immune checkpoint inhibitors. In the US, PCa is the second leading cause of death from cancer in men. The vast majority of men dying of PCa succumb to metastatic castration-resistant disease. PCa frequently exhibits hypoxia. Understanding the role of the TME including hypoxia, metabolism, fibroblasts and the extracellular matrix (ECM), in creating an immune suppressive and metastasis permissive TME can lead to treatments to induce an immune reactive TME responsive to CIT, and reduce mortality from PCa. Imaging methods that provide spatio-temporal information on mechanisms that create barriers to immune cell infiltration to improve the outcome of CIT in PCa are urgently needed to accelerate progress in this field. Here we will apply molecular and functional imaging to understand the role of hypoxia, metabolism, prostate fibroblasts (PFs) and prostate cancer associated fibroblasts (PCAFs) in creating an immune suppressive TME, and in PCa invasion and metastasis. Studies will be performed using human and mouse PCa cells, PFs and PCAFs, preclinical human PCa models, syngeneic PCa models in immune competent mice, and human PCa TMAs. A HIF-1a inhibitor, PX-478, used in clinical trials (NCT00522652) will be used to determine changes in the tumor immune microenvironment and metabolism. In Aim 1, our MR-compatible cell perfusion system that allows careful control of oxygenation will be used to understand the role of PFs and PCAFs in increasing the ability of PCa cells (DU-145, PC-3, RM1 and TRAMP-C2) to invade and degrade ECM under normoxia and hypoxia, and alter metabolism. In Aims 2 and 3 we will use castrate-resistant human PCa xenografts (DU-145 and PC-3) in severe combined immune deficient (SCIO) mice, and syngeneic tumors (RM-1 and TRAMP-C2) in immune competent mice. In Aim 2, in vivo bioluminescent imaging (BLI) of PCa models derived from cells with luciferase under control of the hypoxia response element (HRE) will be combined with PD-L1 PET imaging, ex vivo BLI and second harmonic generation (SHG) microscopy, immunostaining, mass spectrometry imaging and molecular characterization to understand the relationship between tumor hypoxia and and PD-L1 expression, CAF numbers, metabolism, collagen 1 (Col1) fiber patterns and immune cells in these regions. Validation of observations made in tumor models will be performed in human PCa TMAs immunostained for HIF-1a to detect hypoxia and PD-L1/PD-1 and T-cells, a-SMA (smooth muscle actin) to detect CAFs, and SHG microscopy for Col1 fiber patterns. In Aim 3, we will determine the effects of PX-478 on PD-L1 expression, Col1 fibers, metabolism, immune cells, growth, and metastasis. These studies will expand the identification of achievable targets such as hypoxia, CAFs, and metabolism in improving the outcome of CIT in PCa. Justification of animal models In keeping with the NIH emphasis on in vitro systems, we will use our intact cell perfusion system to understand the effects of hypoxia on PFs, PCAFs, and PCa cells on ECM invasion, degradation and metabolism in Aim1. However, the complexity of the tumor microenvironment including the ECM necessitates further validation of these studies with solid tumors in vivo that will be derived from PCa cells engineered to express luciferase under hypoxia for detection with bioluminescent imaging. The use of tumor models will allow noninvasive imaging of hypoxia and PD-L1 to understand the relationship between hypoxia and PD-L1 that will be combined with ex vivo analysis of tumor models and human PCa TMAs in Aim 2. Tumor models will also be required in Aim 3 where we will inject the HIF-1a inhibitor, PX-478, in mice to determine the effects of inhibiting hypoxia in altering the TME and activating the immune microenvironment and reducing metastasis that requires the use of tumor models.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Heart failure (HF) is a major source of suffering, the leading cause of death in the US (higher than dementia and cancer combined) and results in $108 billion annually in global health care costs to individuals, families and communities. Policies limiting reimbursement for HF readmissions increase strain on family caregivers to manage HF in the home with very little support. Despite advancements in guideline-directed medical therapies, HF patient outcomes are worse than the prior decade – including higher readmissions and mortality. Increased uptake of advanced HF surgical therapies including durable ventricular assist device and heart transplantation improve patient health-related quality of life (HRQOL) and survival but require high involvement of family caregivers. The impact of advanced HF care and policy falls on family caregivers who bear increased tasks and responsibilities over time while saving health systems over 7 billion per year. Caregiving for persons with HF, regardless of advanced therapies, is associated with worsening caregiver mental health, burden, stress, loss of social connections, disruptions to work, and increased financial strain. The HEart Failure Resilience Intervention for Caregivers (HEROIC) is a theory- and evidence-based primary palliative intervention including 5 individualized nurse-led sessions conducted over 10 weeks via video calls. HEROIC helps caregivers identify and address caregiving and palliative care needs for the patient, while also focusing on caregiver mental health, leveraging multi-level resilience resources: individual (goal-setting, self-care), community (social support, palliative/clinical, community) and existential (life purpose). We now propose to conduct a mechanistic randomized controlled trial to test HEROIC mechanisms of action for caregivers (N=250) and persons with advanced HF for whom they provide care, including those treated medically, or with ventricular assist device or heart transplantation (N=250) in two health systems. Aim 1 focuses on caregiver self-efficacy and burden as mechanisms of action by which HEROIC improves caregiver mental health at 12 and maintenance at 24 weeks. Aim 2 focuses on patient mechanisms - increases in HF self-care and reductions in unmet palliative care needs as mediators of improvement in patient HRQOL at 12 and 24 weeks and unplanned acute care at 1 year. Aim 3 explores differences in mechanisms and response to intervention components by HF subgroup. We will recruit a Patient and Family Advisory Council to guide the trial and intervention delivery. An exceptional study team with expertise in intersecting fields of research including HF, caregiving, and mixed methods research will guide the study. This innovative proposal directly responds to HF guidelines, policy initiatives and NHLBI strategic priorities, including NOT-HL-23-117 Palliative Care in the Care Continuum, and will provide evidence of a powerful primary palliative care strategy to demonstrate how HEROIC works (mechanisms) so that it can be scaled and adapted across settings and populations.

NIH Research Projects · FY 2026 · 2026-06

Summary This proposal aims to define the vaccine-induced B cell responses, T cell responses, and innate responses that are associated with effective protection against dengue virus (DENV) infection. DENV has become the most important arbovirus worldwide with estimates of ~390 million dengue infections occurring annually, more than 2 million of which are classified as severe disease. Dengue is now endemic in more than 100 countries with approximately 2.5 billion people at risk for infection. Dengue was ranked third in infectious disease threats to the US military, highlighting the need for a dengue vaccine to protect U.S. service members deployed abroad. There are four distinct serotypes of DENV, and the goal of vaccines is to induce balanced immunity against all four serotypes, since unbalanced immunity across serotypes can lead to more severe dengue disease rather than protection. We will study immune responses to vaccines that were tested in a controlled human infection model (CHIM) of DENV infection. These vaccines contained either three of four dengue serotypes, and provided a range of protection against serotype 2 (DENV2) in the CHIM model. The first aim will define the impact of DENV vaccine serotype composition on specificity, function, and durability of serotype- specific memory B cell and monoclonal antibody (mAb) responses and on subsequent responses to DENV2 challenge. We hypothesize that vaccine serotype composition will determine the balance and function of serotype-specific and cross-reactive B cell responses after vaccination, as well as the proportions and neutralizing potency of serotype-specific and cross-reactive mAbs, and that these responses will dictate capacity to effectively prevent DENV2 infection after challenge. In the second aim, we will define the role of T cell function, including metabolic function, on control of challenge DENV2 and on specificity, phenotype and durability of DENV-specific memory B cell and neutralizing antibody responses. We hypothesize that vaccine- induced CD4 T cell phenotype, particularly expression of glucose transporter 1, and function promote effective B and CD8 T cell responses upon DENV challenge, and that CD8 T cell phenotype and function are correlated with control of DENV infection. In the third aim, we will dissect innate immune cell responses to DENV vaccination and challenge. We hypothesize that protective innate cell responses will promote robust CD4 T cell help and facilitate cross-reactive B cell responses. The results from this proposal will define the immune mechanisms underlying effective dengue protection, aid in interpretation of ongoing vaccine trials, and may inform design of next-generation dengue vaccines.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Opioid receptors not only mediate pain relief but also act as the targets of potent exogenous opioids that are devastatingly addictive and cause overdose deaths. The principal target for both analgesia and addiction of exogenous opioids is the µ-opioid receptor (µOR). While traditionally thought to be regulated exclusively by opioids, recent work from our group has detailed an unexpected layer of complexity: endogenous neuromodulators such as endocannabinoids and oxytocin – classically associated with their own independently signaling receptors that regulate mood, pain, inflammation, and social behavior – can directly work to allosterically modulate µOR signaling. This unanticipated ligand-mediated crosstalk opens an entirely new dimension in opioid receptor pharmacology and suggests that targeting these pathways could yield novel interventions for opioid use disorder (OUD) and related conditions. The central goal of this project is to elucidate, at atomic resolution, the molecular mechanisms by which endogenous neuromodulators influence µOR structure and function. We will employ state-of-the-art cryo-electron microscopy (cryoEM) approaches to determine high- resolution structures of µOR in both its active and inactive states, bound to cannabinoids and neuropeptides like oxytocin. Critically, we will advance methodological innovations in time-resolved cryoEM to directly visualize the intermediate conformational states and the full activation/inactivation pathway of the receptor as it transitions between these endpoints. This will allow us, for the first time, to construct “molecular movies” that reveal the stepwise mechanisms by which natural modulators influence receptor activation, allostery, and signaling bias. Through these detailed structural insights, we aim to identify previously unrecognized allosteric sites and intermediate states that serve as "control points" for selective pharmacological intervention. These discoveries will inform the rational design of new therapeutic strategies that harness or mimic the brain’s own modulatory systems, offering a pathway to safer, more targeted treatments for OUD, pain, and overdose. In summary, this project will not only clarify the molecular basis of opioid receptor regulation in the brain, but will also set the stage for mechanism-driven drug discovery to address the current opioid crisis and advance our broader understanding of GPCR biology.

NIH Research Projects · FY 2026 · 2026-06

Heart failure with preserved ejection fraction (HFpEF) is the fastest growing form of heart failure, and is characterized by severe exercise intolerance (EI), exertional fatigue, disability-associated reduced quality of life, and increased mortality. The cause of the severe EI in HFpEF remains unclear, but prior reports and our preliminary data suggest that impaired skeletal muscle energy metabolism may contribute. Previously, our group demonstrated that HFpEF patients experience a rapid decrease in skeletal muscle high energy phosphates (HEP) during exercise, as detected with non-invasive phosphorus magnetic resonance spectroscopy (31P MRS). Other studies found that HFpEF patients have decreased skeletal muscle oxygen delivery and consumption compared to controls. However, due to methodological limitations, it is unknown whether this is due to primary impairments in mitochondrial oxidative metabolism, or whether HFpEF patients have attenuated peripheral blood flow that secondarily limits mitochondrial oxygen utilization. With the need for better in vivo methods to answer this important research question, we recently developed a novel interleaved MRS/MRI tool to simultaneously measure muscle metabolism and peripheral blood flow. Moreover, new clinically available metabolic modulators such as sodium-glucose cotransporter-2 inhibitors (SGLT2i) have been shown to improve clinical outcomes, but their impact on muscle metabolism in HFpEF has not been studied or related to EI. Finally, conventional EI measures during laboratory exercise testing fail to account for activities of daily living or sedentary behavior, but these can now be measured with recent advancements in wearable health technology. However, the relationship between these measures and skeletal muscle energetics have not been investigated in HFpEF patients. Therefore, we will leverage our new MRS/MRI tool to test the central hypothesis that abnormalities in skeletal muscle HEP metabolism are closely linked to manifestations of EI and fatigue in the daily lives of HFpEF patients and can be attenuated with new metabolic modulators. The specific aims are: (1) optimize and refine our novel interleaved MRS/MRI tool and investigate whether rapid HEP decline during exercise occurs despite preserved blood flow in HFpEF patients, (2) explore whether metrics of activities of daily living are closely related to conventional measures of EI and muscle metabolism in HFpEF, and (3) investigate whether SGLT2i administration improves muscle metabolism and reduces EI in HFpEF. The combination of these three elements will give the PI vital experience in developing clinical MRS/MRI research tools, evaluating wearable health device data, and conducting a clinical longitudinal study that will generate crucial preliminary data for a future randomized controlled trial using metabolic modulators. This Pathway to Independence award will be supported by excellent career development resources at Johns Hopkins and mentorship from experts in MR, metabolism, heart failure, wearable technology, and clinical trial design. The new tools and approaches will provide novel insights into EI in HFpEF as well as transferable skills that the PI can leverage in his future research endeavors.

NSF Awards · FY 2026 · 2026-06

The conference "Fundamental Geometries" will be held at the SAIS–Johns Hopkins Institute at the University of Bologna, Italy, on June 22-26, 2026. Geometry plays a central role in mathematics by providing powerful ways to understand complex structures and reveal connections between different areas of science. Recent advances in geometric thinking have uncovered new relationships linking fields such as topology, algebra, combinatorics, and arithmetic geometry, opening fresh perspectives on longstanding mathematical questions. The conference will bring together researchers working in these areas to explore the fundamental geometric structures that underlie many branches of modern mathematics. By promoting the exchange of ideas and collaboration, the meeting will advance basic mathematical research while strengthening international cooperation between U.S. and European institutions. These interactions will expand opportunities for researchers based at U.S. institutions to participate in international collaborations at the forefront of modern mathematics. Support from the project will also strengthen the training pipeline in the mathematical sciences by enabling graduate students, postdoctoral researchers, and early career faculty to engage directly with leading researchers and develop new collaborations. The meeting is intended as the beginning of a continuing series of collaborative events connecting researchers at U.S. institutions with colleagues in Europe. The scientific program will explore emerging directions in modern geometry and closely related fields, highlighting how geometric methods reveal common structural principles across different mathematical frameworks. Topics represented in the meeting include geometric and categorical methods, noncommutative and arithmetic geometry, tropical and combinatorial approaches, and other perspectives that demonstrate the unifying role of geometric thinking. Approximately twenty-five invited talks, together with structured discussion periods, will foster interaction among participants and encourage the exchange of ideas and techniques across disciplines. By bringing together researchers working on complementary approaches to geometric structures, the conference aims to stimulate new collaborations and research directions that may shape future developments in several areas of fundamental mathematics. Conference webpage: https://eventi.unibo.it/fundamental-geometries-jhu-2026 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 · 2026-06

Project Summary Translocation renal cell carcinoma (tRCC) is an aggressive type of kidney cancer lacking effective treatments. It is mostly seen in children, making up 40% of all pediatric and adolescent renal cell carcinomas (RCCs). tRCC is characterized by the chromosomal rearrangements involving the microphthalmia-associated transcriptional factor family, most prevalent of which is the TFE3 transcription factor. Our understanding of the molecular mechanisms of how TFE3 fusions drive tRCC remains limited, and the targeted therapies for these tRCC patients do not exist. In this work, we plan to discover molecular mechanisms by which TFE3 fusion oncoproteins (FOs) drive tRCC, with the goal of identifying potential drug targets and their critical dependencies in the future. Our preliminary data shows that two of the most common TFE3 FOs, NONO-TFE3 and SFPQ-TFE3 form biomolecular condensates inside the nucleus, which are membrane-less compartments organized by weak multivalent interactions. Biomolecular condensates can concentrate pathway-specific factors and change genome organization and transcription programs. Therefore, understanding how NONO-TFE3 and SFPQ-TFE3 condensates form and function can reveal important insights about how TFE3 FOs drive tRCC. Many transcription factors and FOs are known to use intrinsically-disordered regions (IDRs) to form condensates. However, interactions mediated by IDRs are usually ill-defined, hampering the efforts to target these domains in cancer therapy. We found in our preliminary study that highly structured coiled-coil domains (CCDs) and RNA-recognition motifs (RRMs) are important for NONO-TFE3 and SFPQ-TFE3 condensate formation and transcription activities, giving us unique edge to understand how TFE3 FOs form condensates and function. In this proposal, we will investigate further how CCDs and RRMs of NONO-TFE3 and SFPQ-TFE3 FOs drive condensate formation and cancer progression. We hypothesize that TFE3 FOs drive tRCC by forming biomolecular condensates, changing the transcription programs of kidney cells. We aim to: 1) understand how RNA-binding drives TFE3 FO condensate formation; 2) elucidate how TFE3 FO condensates drive tRCC tumorigenesis; and 3) investigate how TFE3 FO condensates reshape the chromatin landscape and transcription of tRCC. Our results will shed light on new therapeutic approaches that target components of FO condensates, their molecular interactions, and their downstream effectors, striving to eradicate the devastating TFE3 FO-driven tRCCs. While we use 2D and 3D in vitro models where possible, we also perform TFE3 FO-driven tumor growth and invasion studies that involve grafting tumor cells into a mouse. This approach is necessary to recapitulate the complex tumor microenvironment that affects tumor progression.

NIH Research Projects · FY 2026 · 2026-06

PROJECT SUMMARY Patients with myeloid neoplasms with loss-of-function TP53 mutations or deletions have a very short overall survival due to lack of effective and safe therapies. Novel approaches with high efficacy and low toxicity are urgently needed. We reported that the atypical chemokine receptor C-C motif receptor-like 2 (CCRL2) is overexpressed in hematopoietic progenitor cells from patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) arising from MDS and that its deletion suppresses MDS/AML cells growth and sensitizes them to hypomethylating agents making CCRL2 an attractive target in myeloid neoplasms. We discovered that TP53 mutated MDS/AML samples express the highest levels of CCRL2 across AML subtypes. Thus, we developed an anti-CCRL2 antibody-drug conjugate (ADC), which shows significant anti-leukemic activity against TP53 mutated MDS/AML cell lines and primary samples in vitro and in cell line- and patient-derived TP53 mutated AML xenograft models without any effect against healthy hematopoietic cells and systemic toxicity in healthy mice. Moreover, screen of 171 anti-cancer agents that are either FDA approved or under investigation in clinical trials revealed that the PARP1/2 inhibitors talazoparib and stenoparib have the highest synergistic effect when combined with the anti-CCRL2 ADC against TP53 mutated MDS/AML cells. However, additional patient-derived xenograft studies and assessment of the efficacy and safety of anti-CCRL2 ADC in an immunocompetent TP53 mutated syngeneic AML model are needed for the validation of the anti-leukemic activity and safety of this agent before its transition to early phase clinical trials. Moreover, the synergistic or additive effect of the addition of PARP1/2 inhibitors to anti-CCRL2 ADC need to be validated in in vitro and in vivo studies. In the first aim of this study, we will analyze the efficacy of recurrent doses of anti-CCRL2 in patient-derived TP53 mutated MDS/AML xenografts following a successful approach engrafting these samples and in lethally irradiated CD45.1 recipients of bone marrow cells from Jak2V617F/+ Trp53−/− and Jak2V617F/+ Trp53+/− mice. In the second aim of this study, we will analyze the additive or synergistic effect of combining the anti-CCRL2 ADC with the PARP1/2 inhibitors talazoparib and stenoparib both in vitro by treating TP53 mutated MDS/AML cell lines and primary samples and in vivo using our established cell-line derived TP53 mutated MDS/AML xenograft models. Murine models are scientifically justified because they permit evaluation of the anti‑CCRL2 ADC and its combination with PARP1/2 inhibitors in vivo using both xenograft and immunocompetent TP53‑mutated AML models that closely recapitulate human myeloid neoplasm biology and drug responses, which cannot be achieved in vitro or with non‑mammalian systems. These studies are essential to establish efficacy, define on‑ and off‑target toxicity, and generate the preclinical data required by NIH and regulatory agencies before initiating early‑phase clinical trials in patients with TP53‑mutated myeloid neoplasms. Our studies have the potential to introduce a new targeted therapy with low off- and on-target toxicity and a novel combinational strategy that can improve the remission depth urgently required for patients with TP53 mutated myeloid neoplasms to achieve longer survival.

NIH Research Projects · FY 2026 · 2026-05

PROJECT SUMMARY The antisense gene asp maps in the HIV-1 genomic region overlapping env at the SU/TM boundary. Asp is found in pandemic strains of group M, but not in other primate lentiviruses including non-pandemic HIV-1 groups N, O, and P. We showed that asp is highly conserved despite constraining the evolution of env. We also reported that the asp gene is found at a higher frequency in people living with HIV-1 (PLWH) who progress to AIDS in <3 years (rapid progressors) compared to those who progress to AIDS in >12 years (long term non-progressors). The asp gene encodes the 189-aa hydrophobic protein, ASP. We reported that ASP shows high sequence iden- tity across HIV-1 isolates from all group-M subtypes. Work from our lab described the presence of ASP on the plasma membrane of infected cells, and on the envelope of infectious HIV-1 particles. Our unpublished studies demonstrate that the presence of ASP on the surface of HIV-1 particles facilitates viral entry. Several studies have shown the presence of cellular and humoral immune responses to ASP in PLWH, which proves its expression in vivo. A recent report reported that antibodies against ASP were specific for epitopes in the predicted ectodomain of ASP. Our preliminary studies confirmed the presence of antibodies against the ASP ectodomain in Elite Controllers (EC). Yet, none of the studies published so far endeavored to isolate ASP anti- bodies from PLWH and to test their functional activity as a way to investigate the role of ASP in HIV-1 infection. The overall aim of this application is to isolate monoclonal antibodies (mAbs) against the ectodomain of ASP from EC, Viremic Controllers, and PLWH both on and off ART. We will test their activity in in vitro and ex vivo assays. These studies will be performed in collaboration with Dr. Mohammad Sajadi (Institute of Human Virology, University of Maryland School of Medicine), who has established a cohort of >200 PLWH from whom he has already obtained paired serum and PBMC samples that are immediately available for the studies proposed here. Dr. Sajadi has developed a method for the identification, isolation, and cloning of mAbs that led to the discovery of best-in-class mAbs against HIV-1, SARS-CoV2, and CCHFV. Here, we propose the following specific aims: In Specific Aim 1, we will generate pools of overlapping peptides that span the ectodomain of ASP, and we will use these peptide pools to screen serum samples from PLWH in Dr. Sajadi’s cohort to identify those with strong- est binding to each of the five ASP peptide pools, and to determine their peptide sequence specificity. Next, we will use single-cell PCR and mass spectrometry to isolate and clone high affinity anti-ASP mAbs from the paired PBMC samples of the same donors. We will then validate the ASP specificity of these mAbs in ELISA and virion capture assays. In Specific Aim 2, we will test the activity of the ASP mAbs in mediating antibody dependent cellular toxicity (ADCC), reducing viral entry in single-round infection and viral replication in multiple rounds of infection, and detecting ASP on the cell surface, in the cytosol, and within nuclei.

NIH Research Projects · FY 2026 · 2026-05

PROJECT SUMMARY Memory consolidation is an important process in memory retention, whereby short-term memories are transformed into long-term ones. Memory retention is facilitated when novel or salient experiences occur within 1-2 hours before or after memory encoding. Novelty-induced memory consolidation relies on the locus coeruleus-hippocampus pathway and involves the de novo synthesis of an–as–yet–unidentified group of proteins known as plasticity-related proteins (PRPs) in the hippocampus. These PRPs are thought to play a critical role in memory consolidation by stabilizing synaptic plasticity. However, the precise mechanism through which these newly synthesized PRPs facilitate the consolidation of synaptic plasticity and memory in the hippocampus remains unclear. This uncertainty primarily arises from the limited availability of loss-of-function techniques capable of perturbing the function of the target PRP with a minute temporal resolution and selectively within this memory process. This project aims to develop innovative molecular tool, using chemical-based manipulation, that enable the rapid and precise inactivation of endogenous target proteins in specific cell types during memory consolidation without genetic modification. Specifically, upon a chemical trigger, our new tool is designed to relocate a target PRP from its active site to other subcellular locations within a timeframe of minutes. This relocation can be achieved in specific neuron types in the hippocampus of freely moving rats undergoing memory tasks. To demonstrate the utility of this tool, we have selected activity-regulated cytoskeleton associated protein (ARC), one of the major PRPs, as a model target. Success in this endeavor will not only shed light on the elusive molecular mechanisms of memory consolidation by identifying key PRPs but also potentially reveal new pharmacological targets to improve memory retention. Moreover, the flexible and modular nature of this genetically encoded tool opens the door to its use in a broad range of biological applications beyond memory research.