University Of Chicago

NSF Awards · FY 2026 · 2026-10

Cloud computing is essential to a growing number of science use cases, but configuring scientific environments for deployment in the cloud can be challenging given that it often requires specialized knowledge in system administration, networking, security, and involves numerous configuration settings. Furthermore, many scientific workloads depend on tightly coupled virtual clusters, specialized hardware, fast interconnects, accelerators, and custom drivers. Managing this complexity consumes valuable time and attention that researchers could otherwise devote to science. This project develops an AI-based conversational assistant for configuring scientific computing environments that lets researchers describe what they need in everyday language and then creates the environment and verifies its integrity on shared research cloud infrastructure. The benefits of this approach range from increasing scientific productivity and lowering the cost of using cloud computing to enabling practical reproducibility of computational experimentation. The project designs and deploys an AI-based agent framework that can plan, provision, and validate scientific computing environments on open research computing infrastructure, such as Chameleon and Jetstream2. The framework combines large language models running on open, high-performance academic hardware with a set of software tools exposed through standard interfaces that include cloud-based services for resource provisioning, hardware and software environment templates, correctness checks, as well as validation benchmark suite. Key components include planning modules with built-in checks on resource limits, timing, and hardware compatibility; state and error handling modules that track multi-step workflows and summarize system events; and search pipelines that organize information from a wide variety of sources, including documentation, logs, help desk tickets, and environment artifacts into a searchable knowledge base. 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 · 2026-09

The project aims to advance the field of statistical change-point detection by developing novel methods and their associated theory to handle complex data with irregular signals. Unlike the traditional setting in which signals before and after the change point are often assumed to differ by a constant shift, irregular signals refer to the situation when the post-change signal may vary in highly unpredictable ways without any pre-specifiable pattern or structure. This can pose a tremendous challenge on many existing change-point tests, often resulting in notable reductions in their statistical power and increasing their vulnerability to maliciously designed adversarial attacks. By allowing the post-change signals to be irregular and not necessarily follow the standard assumptions as in conventional change-point analyses, the research developed in this project is expected to lead to more robust and next-generation statistical and machine learning protocols and toolboxes with rigorous theoretical guarantees for change-point detection in a wide range of applications. For example, detecting abrupt changes in power grids, attacks in sensor networks, or emerging trends in social networks all require powerful methods for detecting irregular changes. As a result, the research will advance not only the field of statistics but also a range of other disciplines including machine learning and artificial intelligence where data with irregular signals may arise. The research will also be integrated into the undergraduate and graduate education at participated institutions to equip students with advanced yet accessible statistical and machine learning knowledge for analyzing data with irregular signals. The research involves the development of novel statistical methods and their associated theory for change-point detection and estimation in the presence of potentially irregular signals. To quantity the uncertainty in the estimated signals from dependent and noisy data, a causal representation framework is employed with a suitably constructed functional dependence measure to quantify the effect of dependence via the technique of perturbation and innovation coupling. This enables the use of deep probabilistic tools, such as the invariance principle and Gaussian approximation results, for a general class of dependent processes to guide the selection of a statistically appropriate alarm threshold for detecting change points in the presence of irregular signals. The project aims to address change-point detection under irregular signals both in the offline setting, where the analysis is performed after all the data are collected, and in the online setting, where sequential testing becomes desirable as data arrive. In addition, different asymptotic schemes are considered to address situations in which stable historical data are available and when such data are not available to practitioners. The research is also expected to promote scientific and technological advances in applications that require rapid anomaly detection with complex alternatives. 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 · 2026-08

This award supports a summer workshop for graduate students and early-career scientists that combines lectures with hands-on small groups research projects in hackathon-style format. The workshop addresses the growing need for more accurate forecasts of conditions that affect daily life, from severe storms to heat waves to long-term shifts in weather patterns. Reliable forecasts across a wide range of time scales are essential for protecting life and property, and supporting preparedness and planning. By bringing together emerging researchers, the workshop also helps build the skilled workforce needed to advance forecasting science. The theme of this year's workshop is understanding the Earth systems across scales, motivated by the need for better predictions of weather phenomena. Participants will explore several key challenges, including why forecasts often lose accuracy in the period between about two weeks and one month; how interactions across scales, from local weather systems to broader environmental patterns, can be better represented in forecast models; and how recent advances in artificial intelligence and machine-learning approaches are enabling new forecasting systems that can match or even surpass the performance of traditional 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 · 2026-08

Artificial intelligence is increasingly used to guide scientific discovery, engineering design, and complex decision-making, where each experiment or trial can be costly and time-consuming. A central challenge is how to efficiently identify the most informative experiments from vast and complex design spaces, especially when observations are limited and uncertainty is high. This project develops a new paradigm for adaptive experimental design that enables learning systems to not only model data but also actively decide what data to acquire. The project's novelties are the integration of data representation and experiment selection into a unified learning framework, where the strategy for choosing experiments is itself learned from data rather than specified by fixed rules. The project's broader significance and importance are in accelerating scientific discovery, improving the efficiency of engineering systems, and enabling intelligent decision-making in settings where data collection is expensive or constrained. Technically, the project formulates adaptive experimental design as a coupled optimization problem that jointly learns representations of experiments and policies for selecting new measurements. It develops learning-based acquisition strategies using tools from representation learning, probabilistic modeling, and sequential decision-making. The approach includes methods for uncertainty-aware modeling in high-dimensional settings, architectures that learn to prioritize informative data points, and algorithms that leverage simulation and historical data to train decision policies. It further incorporates multi-fidelity data sources, indirect feedback, and parallel experimentation into a unified framework, enabling scalable and robust decision-making in complex environments. The resulting system is evaluated in applications such as scientific simulation, cyber-physical system optimization, and data-driven protein design, demonstrating improved efficiency and adaptability. This work advances the foundations of data-driven discovery and enables broader adoption of AI in real-world experimental systems. 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 · 2026-06

Three dimensional models power modern technologies ranging from engineering and manufacturing to robotics and digital content creation, yet creating and modifying these models remains a difficult and time-consuming process even for experts. Even basic operations, such as selecting or editing parts of a shape, require complex tools and significant effort. As a result, scientists, engineers, and creators cannot easily generate or manipulate 3D content. This project seeks to remove these barriers by developing artificial intelligence technologies that fundamentally transform how people interact with 3D shapes. By enabling users to create and edit 3D objects through intuitive interactions such as language, images, and simple geometric selections, the project aims to make 3D modeling accessible to a much broader community. These capabilities will accelerate innovation across the many fields that depend on geometric data. The project develops new computational foundations for artificial intelligence systems that operate directly on meshes, the dominant representation of geometry in graphics, engineering, and simulation. The research will first establish a mesh foundation model that learns general geometric representations capable of supporting tasks such as segmentation, correspondence, and analysis across diverse shapes. Building on these representations, the project will develop multimodal algorithms that allow users to edit and transform meshes using text descriptions, images, point selections, and other intuitive inputs while preserving geometric structure. Finally, the research will investigate the use of large language models as generative agents that can synthesize, refine, and evaluate 3D data, enabling scalable pipelines for generating training data and accelerating progress in 3D artificial intelligence. Together, these advances will provide new tools for understanding and creating geometric data while training students and researchers in emerging areas of artificial intelligence. 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 · 2026-06

Modern artificial intelligence systems produce striking images, text, and other creative outputs, but it is often unclear what these systems have actually learned internally. This makes it difficult to ensure that these models are reliable, safe, and trustworthy when deployed in the real world. Although these models can imitate patterns in data, the process through which they do so does not necessarily correspond to meaningful causes, stable mechanisms, or interpretable concepts that stakeholders can decipher and diagnose. This project develops a new statistical framework for building AI models designed to uncover interpretable, generalizable structures hidden inside complex, high-dimensional data such as images, language, and scientific measurements. This research investigates the statistical foundations of generative AI. A key goal is to understand how and when generative models learn reusable, causal structure and what the tradeoffs are. Specifically, the project focuses on understanding how generative models learn complex, high-dimensional structures without suffering the curse of dimensionality and how they can learn interpretable causal factors from data. This will deliver a framework with practical models and algorithms for reliable generative AI that is both independently verifiable and reproducible. The work will combine ideas from causal inference, nonparametric statistics, latent variable modeling, and deep learning to develop methods with rigorous guarantees. The goal is to move beyond black-box imitation toward AI systems whose internal factors can be interpreted, tested, and used to understand how complex systems change. 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

The International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI) is the major international conference in medical image computing, image-guided interventions and robotics. It promotes, preserves, and facilitates research and education, and its proceedings foster the exchange and dissemination of advanced knowledge by leading institutions, scientists and physicians. The conference is the result of the joint efforts of previous major conferences in three fields: Visualization in Biomedical Computing (VBC), Computer Vision and Virtual Reality in Robotics and Medicine (CVRMed), and Medical Robotics and Computer Assisted Surgery (MRCAS). Since its first edition in 1998, it has become the premier conference in the field with its proceedings having citation scores comparable to high-impact journals. Conference topics include, computer vision & image processing in medicine, computer-aided diagnosis, computer-assisted interventions, guidance systems & robotics, visualization and virtual reality, biomedical imaging applications, and imaging systems, spanning disciplines such as radiology, pathology, surgery, oncology, cardiology, physiology, and psychiatry. MICCAI includes three days of oral presentations, poster sessions and invited keynote talks. All paper submissions undergo a rigorous double-blinded peer-review (~30% acceptance) and several papers have become landmark publications with thousands of citations. Satellite events and educational initiatives with similar attendance rates to the main conference also take place on the day before and after the conference in the form of workshops, tutorials and challenges. MICCAI span the entire globe and rotates every year between three geographical zones: the Americas, Europe/Africa/Middle East, and Asia/Oceania. Attendees from dozens of countries typically have a strong student representation (40-50% in the last editions). This proposal requests to support the attendance of US-based students and early career investigators through travel awards to support their education, training and networking opportunities. The supported trainees will be able to attend the conference to learn from the latest advances in the field, participate in the MICCAI Mentorship Program to enhance their career development, benefit from networking opportunities and participate in educational events and sessions including tutorials, workshops and challenges.

NSF Awards · FY 2026 · 2026-06

This award provides an opportunity for U.S. early career members to present at this prestigious conference. The DPF meetings are the largest gathering of the U.S. particle physics community, and they include substantial international participation. The 2026 DPF meeting will take place on July 20-24, 2026 at Fermilab. Notably, the DPF meetings provide an important venue for junior physicists to present their research works, to interact with senior, highly respected physicists, and to nurture their career development. The DPF 2026 will gather active researchers of all senior and junior levels together to discuss a broad range of topical issues in particle physics, recent developments in particle theory, future directions in high-energy experiments, and recent developments in accelerator, detector, and computing technologies. The meeting stimulates new ideas and encourages collaboration for future research. The DPF meetings bring active researchers from related fields such as nuclear physics, astrophysics, atomic physics, and computational physics, stimulates innovative ideas and encourages multi-disciplinary collaboration. They also offer an excellent venue for engagement of our scientific community as well as the general public in broader impacts through public lectures and outreach programs. 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

ABSTRACT Proprioception is the sense that allows animals to monitor their body position and movements; proprioceptive deficits lead to severe challenges in moving and maintaining posture. Proprioceptive feedback passes through local pathways targeting motor circuits, where proprioception adjusts ongoing movements, as well as projection pathways to the brain, where proprioception is used to learn and plan future actions. The fundamental differences between projection and local pathways remain largely unknown. Our long-term goal is to understand proprioceptive circuit activity during natural animal behavior, with a focus on the roles played by genetically defined cell types. The objective of this proposal is to characterize the fundamental differences between projection and local pathways. We focus on second-order neurons, CNS neurons that receive direct input from proprioceptive sensory neurons. We will specifically address the following questions: Does the brain receive minimally processed proprioceptive stimulus information or integrated representations of specific stimulus features? Is information presented to the brain in a behaviorally state-dependent manner? Does the brain receive different types of information in comparison to local circuitry? We use Drosophila larvae as a highly tractable model to study proprioception. This proposal leverages two major technical innovations: (1) CRASH2p microscopy, which enables volumetric imaging of neural dynamics in intact, freely moving, and behaving larvae, and (2) connectomics, which facilitates the comprehensive reconstruction of synaptic connections between second-order neurons and their synaptic partners. Based on preliminary data, we will test the central hypothesis that local and projection second-order proprioceptive neurons differentially integrate and process naturally occurring self-movement stimuli. We test this hypothesis using complementary in-depth (functional, Aim 1) and in-breadth (anatomical, Aim 2) approaches. The proposed research is significant because it will provide two advances that, to date, remain out of reach in other models. First, it will provide a comprehensive anatomical understanding of the full complement of second-order proprioceptive neurons and the networks in which they are embedded. Second, it will produce first-of-its-kind knowledge of the activity of second-order proprioceptive neurons in intact animals performing multiple behaviors and determine the role of a specific type of proprioceptor in shaping that activity. Thus, our work is expected to provide a new conceptual framework for understanding how various second-order neurons integrate and process proprioceptive information, as well as how the brain senses proprioceptive stimuli.

NIH Research Projects · FY 2026 · 2026-06

Project Summary Deciphering how DNA sequences encode regulatory activities provides a new lens for dissecting the underlying molecular mechanisms of genome regulation. However, although deep learning approaches have significantly advanced our ability to predict chromatin accessibility, transcription factor binding, 3D genome architecture, and transcription from DNA sequence, current models remain limited in their ability to deliver mechanistic insights, cannot effectively capture long-range sequence dependencies, and do not reflect the heterogeneous and dynamic nature of molecular-level genome regulation. Motivated by our observation that simple explainable models can simultaneously contribute to improved biological understanding and predictive power— indicating that we are still in a data-limited regime—this proposal aims to address these challenges by developing a new generation of sequence-based AI frameworks focusing on biology-informed simplicity, explainability, and mechanistic insights without compromising on performance. Our approach is built on three key pillars: (i) We will develop explainable sequence models that discover simple rules for diverse regulatory functions. (ii) We will develop novel architectures to accurately predict long-range regulatory interactions directly from sequence, and generate genome-wide regulatory interaction maps across cell types. (iii) We will pioneer "sequence-to-distribution" generative models to capture the dynamic, single-molecule nature of chromatin regulation, unveiling its inherent heterogeneity and the sequence determinants of molecular cooperativity. Achieving these goals promises both improved predictions and deeper mechanistic insights, ultimately contributing toward making genomic regulation an "open book" where every single base’s contribution can be explained.

NIH Research Projects · FY 2026 · 2026-06

Abstract A major goal in biomedical science is to move beyond static images of proteins and other biological macromolecules to the internal dynamics underlying their function. This level of study is necessary to understand how these molecules work, to engineer new functions, and to rationally develop therapeutics. Work in the past few years has finally enabled new nascent technologies for observing intramolecular dynamics in proteins and other biological macromolecules. In this application, we propose to advance a particularly exciting approach called Electric-Field Stimulated X-ray crystallography (EFX). EFX involves the use of large external electric fields to induce subtle motions within proteins and other biological macromolecules and to simply watch the resulting motions with atomic resolution and over a range of timescales in which motions are functionally relevant. This method has the potential to map protein mechanisms with unprecedented detail, but requires significant further work to broaden its application and to make it accessible. In this proposal, we describe the key advances – hardware, software, and computational – that can achieve these goals. We will install and test new machinery to generalize the method and will calibrate the size and orientation of the forces applied by electric fields at local environments within molecules. We will develop a modern, open-source, end-to-end software pipeline for reduction and analysis of raw data and will simplify the process of making biological interpretations. Finally, we will advance computational simulations that can work together with experiments in a virtuous cycle to fill in and extend the experimental analysis. Taken together, the work proposed here will enable the scientific community to practically use of the EFX method as a general tool for understanding proteins and other complex biological macromolecules.

NIH Research Projects · FY 2026 · 2026-06

Project Summary / Abstract The University of Chicago Integrated Light Microscopy Facility (ILMF) requests funds to purchase a high-end laser scanning confocal microscope. The ILMF currently serves 420 users in 80 labs from across the University. Sixty-seven of those labs use laser scanning confocal microscopy, and 78% of those labs have NIH funding. Usage hours have increased as the ILMF’s microscope capacity has decreased. Two of our confocal microscopes, both Leica SP5 models, are over 14 years old. Leica has designated them end-of-life, meaning they are no longer manufacturing parts for these systems and replacements are not guaranteed. We have already experienced failure of the 488nm Argon and 592nm depletion lasers on one, and failure of the Mai Tai multiphoton excitation laser on the other, with no possibility of replacing any of these components. We expect to decommission at least one SP5 within the next year, making users hesitant to start new projects on those systems. This has stressed our two newer laser scanning confocal systems (purchased with institutional funds in 2016 and 2020), pushing them to use levels averaging 91% of AUT, defined as 3640 hours per year. The system proposed here is the Evident (formerly Olympus) Fluoview 4000 (FV4000), released in 2024. The system will increase the capacity and functionality of laser scanning confocal microscopes in the ILMF, allowing users to collect high-quality data more readily. Several features of the FV4000 will be new to the ILMF, and satisfy a number of outstanding investigator needs. Features include: state-of-the-art, patented, fast signal processing silicon photomultiplier (SiPM, Evident SilVIRTM) detectors, to significantly improve signal-to-noise levels, enhancing detection of Golgi cisternae and other organelle sub-structures; four high magnification, long working distance silicone immersion objectives for detailed, multi-color, 3-dimentional imaging of organoids, thick tissues and tumor samples; and three near-infrared wavelength lasers for excitation of fluorophores beyond the current imaging spectrum, allowing for investigation of a larger number of molecules of interest in a single sample. The FV4000 will also feature full environmental control, allowing users to take advantage of faster imaging speeds to image live samples. This will make it possible to image longer sessions at higher frame rates with less photodamage, resulting in more robust and reliable data from live samples than currently possible. Finally, the FV4000 base is modular in design, allowing for field upgrades with Evident or third-party resources (e.g. a single molecule localization module) as users’ experimental needs grow. In summary, adding an Evident FV4000 laser scanning confocal microscope to the ILMF will make it possible for users to gather information from samples that are currently challenging but valuable research models.

NIH Research Projects · FY 2026 · 2026-06

Project Summary: Preclinical and clinical evidence suggest that radiation therapy (RT) and CAR-T therapy have complementary and/or synergistic effects through multiple possible mechanisms including increased CART infiltration, induction of target protein expression, and activation of innate immunity. Our novel radioinducible plug-and-play CAR-T system could potentially transform the clinical use of radiation, gene therapy, and CAR-T therapy in solid cancers by improving local tumor control and treatment of metastatic disease. RT has the potential to increase the efficacy of CAR-T therapy by debulking the tumor, restricting the need for toxic systemic therapies, promoting tumor infiltration by CAR-T, increasing the expression of target molecules by cancer cells, and amplifying the native immune response. All these mechanisms address current obstacles in CAR-T therapy of solid tumors. In Aim 1 we will generate synthetic radioinducible promoters, CAR-T production protocols, and RT/drug combinations that result in optimized transgene induction and CAR-T function in vitro. In Aim 2 we will use irradiation of tumors (s.c. and orthotopic) in mice to activate cytokine production by tumor-infiltrating radioinducible (Rad-I) GA1CAR-T cells, and to increase antigen expression, taking advantage of the ability to switch targets in a plug-and-play fashion provided by the GA1CAR. We will also leverage IR-induced changes in protein expression to explore novel IR-induced targets. In Aim 3 we will study the antitumoral mechanisms of radioinducible and switchable Rad-I GA1CAR-T therapy combined with IR, with a special focus on the prevention of tumor antigen loss variant (ALV) escape , using both xenograft and immunocompetent mouse models. We will also image the destruction of antigen-positive and negative cells, by longitudinal in vivo microscopy using fluorescently labeled cancer cells, CAR-T cells, and tumor vasculature, in tumors treated with the Rad-I plug-and-play GA1CAR-T system and IR. The effects of anti-cancer therapies must be tested in animal models because in vitro models are inadequate for translation to the clinic; mouse models are the primary source of pre-clinical data for human phase I trials. Discovery of novel therapeutic targets in our proposal depends on the positive or negative selection of genetically modified cancer cells in an immune-dependent fashion. Selection is influenced by the complex and dynamic interactions between modified cancer cells with mouse immune system that can only be elicited in vivo. These so-called “tumor microenvironments” are impossible to recreate in vitro or in silico with current technology. The proposed studies will require the use of both immunocompetent and immunodeficient mice. C57BL/6 mice are chosen because GL26, SB28 and LLC tumors grow in this mouse strain. NSG are selected because human cancer cell lines (SKBR-3, HCC1954, SKOV-3, SH-SY5Y, A549 and U87) grow in them. NSG-A2 mice are selected because they express HLAA2 human MHC-I molecules and therefore can be used to assess graft-vs-host responses.

NSF Awards · FY 2026 · 2026-06

Visualization facilitates data communication across the sciences and society at large, yet it can be hard to know whether a visualization gives an honest depiction of evidence. One reason for this is that visualizations provide necessarily incomplete views on complex datasets. Charts that attempt to convey too much information become incomprehensible, so honest and effective visualization design requires authors to choose what information to disclose and, conversely, what aspects of data will be hidden or distorted. This project will address the problem of responsible data disclosure through visualizations. For visualization authors, it will build tools that help them balance goals such as effective communication and protecting the privacy of data subjects. For audiences, it will develop new ways to support skepticism about what a chart cannot show by design. The project will also produce novel educational materials and games to help students learn to use visualizations responsibly and avoid misinterpretations. Together, these activities will create a practical framework for understanding and espousing ethical standards for data communication. This project reframes visualization as a mechanism for data disclosure. It develops a theory defining visualization design goals in terms of balancing forms of information loss that designers and audiences care about. The theory makes these losses computable by grounding them in mathematical formalisms developed through analysis of examples, synthesis, and expert interviews. Codifying this formalism in software will enable automated reasoning over the space of possible visualization designs suited to a given goal. Indexing this design space on relevant forms of information loss will enable new ways to recommend solutions for visualization authors, as well as new interfaces that generate assistive explanations for audiences. These tools will be evaluated through software testing, user studies, and controlled experiments. To teach students about ethical data communication, and engage them in research, this project will develop data disclosure games that create opportunities to practice and reflect on responsible ways of navigating information loss with visualization. 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 · 2026-05

PROJECT SUMMARY/ABSTRACT Inflammatory bowel disease (IBD) is characterized by chronic relapsing intestinal inflammation with unpredictable flares that significantly impair patient quality of life. The gut microbiome plays a central role in IBD pathogenesis, yet specific mechanisms linking gut bacteria to inflammatory processes remain poorly defined. Glucocorticoids function as critical mediators in IBD both as endogenously produced anti-inflammatory molecules that help maintain mucosal homeostasis and as exogenously administered therapeutic agents for managing acute flares. This proposal investigates a novel bacterial pathway that enzymatically converts active glucocorticoids to inactive metabolites in the gut microbiome. Our preliminary data demonstrate that steroid- metabolizing bacterial enzymes are significantly enriched in IBD patient microbiomes compared to healthy controls, suggesting a potential contribution to disease pathogenesis. We hypothesize that microbial inactivation of glucocorticoids disrupts anti-inflammatory signaling and exacerbates intestinal inflammation. To test this hypothesis, we will employ gnotobiotic mouse models with genetically defined bacterial strains that either possess or lack glucocorticoid-metabolizing capability, evaluating their impact in the presence or absence of therapeutic glucocorticoid treatment. This exploratory research aims to investigate a potential novel mechanistic link between microbial modulation of host endocrine signaling and inflammation, which may ultimately inform the development of microbiome-based diagnostics and therapeutic approaches for inflammatory digestive diseases.

NIH Research Projects · FY 2026 · 2026-05

PROJECT SUMMARY BCL-2 family of proteins are well-known regulators of cellular apoptosis. Residing at the mitochondria, this family is made up of pro-apoptotic and anti-apoptotic partners whose interactions are regulated by BH3-only proteins. Designed to mimic BH3 proteins and dissociate BCL-2 binding partners to induce cancer cell death, there continues to be discovery of BH3-mimetics that target a range of BCL-2 anti-apoptotic proteins. However, very little is known about how these therapeutics induce BCL-2 family protein:protein interactions (PPIs) changes in other cells, particularly immune cells which heavily rely on the BCL-2 family during ontogeny, homeostasis, activation, and signaling. New evidence demonstrates that targeting of certain anti-apoptotic proteins modulates immune functions via unknown mechanisms distinct from apoptosis. Therefore, there is clinical need and timely opportunity given their increased use to maximize BH3 mimetics through defining the mechanisms of action responsible for these changes and to determine how BCL-2 family targeting affects discrete immune cell populations, causes immune system adaptation, and alters effector functioning. Our long-term goal is to dissect non-apoptotic immunomodulatory effects governed by BCL-2 proteins in T cells and determine how drugging this system with BH3 mimetics affects their function for ultimate translation into T cell-based therapies. Thus, the overall objective of this proposal is to mechanistically characterize the functional consequences of BH3 mimetic therapy on human T cells used for adoptive therapy and discover how BCL-2 PPI alterations control T cell effector function, plasticity, and mitochondrial dynamics. We propose that these studies will significantly broaden the clinical translatability of this drug class and identify novel druggable immune modulatory mechanisms. Based on our preliminary findings, we hypothesize that BCL-2 family anti-apoptotic protein PPI inhibition with venetoclax and other BH3 mimetics will directly alter the effector function of adoptive T cell therapy produced from cells isolated from healthy donors and heavily pretreated patients through adaptive apoptotic, genetic, metabolic, and functional changes. These hypotheses will be tested by pursuing two broad specific aims: 1) Test the potential of BH3 mimetics to alter adoptively transferred T cell function using chimeric antigen receptor (CAR) T cells as a model system and 2) Interrogate PPI alterations within the BCL-2 family interactome at baseline and following BH3-mimetic treatment in CAR T cells to determine the mechanism(s) of action responsible for their immunomodulatory effects. We believe the research proposed in this application is innovative because it seeks to expand the discovery of what we understand about BCL-2 family control of T cell function and determine how BH3 mimetics can be used to modulate T cell activity. We believe this research is of great significance given the growth of BH3 mimetics in treating a wide range of diseases and will provide justification for using this drug class as bona-fide immunomodulatory agents.

NIH Research Projects · FY 2026 · 2026-05

Summary My laboratory investigates how germ cells distinguish self from non-self RNAs to safeguard genome integrity across generations. Central to this work are small RNA pathways and the specialized condensates known as germ granules, which coordinate RNA regulation in the C. elegans germline. These mechanisms are crucial for protecting the genome from transposons and inappropriate gene silencing, ensuring fertility and proper development. Over the past decade, we have defined key rules that govern piRNA targeting and established how gene licensing pathways, such as CSR-1–associated small RNAs, actively protect self RNAs from piRNA-induced silencing. Our research has shown that licensing and silencing are not simply antagonistic but rely on highly orchestrated sorting decisions made within distinct sub-compartments of germ granules. We have also developed innovative tools, including synthetic piRNA systems and fluorescent reporters, that allow us to dissect small RNA activity and track regulatory outcomes in vivo. Building on these advances, our research over the next five years will address how intrinsic RNA features and associated proteins direct RNAs through specific regulatory fates—either protected or silenced. A major focus will be to understand how the spatial organization of germ granules, including functionally distinct sub-compartments, contributes to these decisions. We aim to elucidate the molecular mechanisms by which licensing pathways restrict RNAs from entering piRNA silencing domains, and how condensate interfaces regulate RNA flow and protein sorting. To achieve these goals, we are leveraging a suite of cutting-edge tools, including live- cell imaging, proximity labeling, CRISPR-based reporters, and inducible perturbation systems. These approaches will allow us to visualize dynamic RNA-protein interactions in space and time and to manipulate regulatory pathways with precision. Ultimately, our work seeks to uncover fundamental principles of small RNA-based genome defense, condensate organization, and epigenetic inheritance. These insights have broader implications for RNA regulation in other systems, and may inform the development of gene regulatory and therapeutic strategies that harness small RNAs for targeted gene regulation.

NIH Research Projects · FY 2026 · 2026-05

Project Abstract Among the thousands of genetic disorders in humans, most are rare and lack effective treatments. Suppressor tRNA therapy has come a long way to become a realistic treatment option for genetic diseases. Sup-tRNAs read through premature stop codons derived from genetic mutations in translation to generate full-length proteins. Compared to gene or mRNA-based therapies that deal with one gene or one mutation at a time and require safety and efficacy assessment for each therapeutic agent, a major advantage of tRNA therapy is the potential of using the same tRNA to treat many diseases that just share a common genetic mutation in many genes. However, about 30% of human genetic disorders are from missense mutations that cannot be treated with sup- tRNAs. For those disorders, missense-correcting tRNAs are needed which are charged with one amino acid but read the codon of another amino acid in translation. PI's lab has developed a missense-correcting tRNA identification platform and applied it to identify such tRNAs charged with Arg but read Gln/His/Trp/Cys codons which correspond to most frequent missense mutations in genetic disorders. Aim 1 will develop a high-throughput engineering platform for missense-correcting tRNAs that will be more efficient, and at the same time, less toxic to cells. We will screen tens of thousands of potential missense-correcting tRNAs simultaneously in cellular models. Aim 2 will screen for missense-correcting tRNA expression with reduced toxicity to cells. Our studies will establish a foundation for using missense-correcting tRNAs as a new therapeutic modality for rare diseases.

NSF Awards · FY 2026 · 2026-05

The Euler and the Navier-Stokes equations are the most established models in fluid dynamics. Scientists and engineers apply these equations to model various phenomena, including weather patterns, ocean currents, flows around vehicles, aircraft, and ships, as well as blood flow. Mathematicians and physicists believe that understanding the solutions to these equations can lead to an explanation for turbulence. Despite their wide range of applications, there is no theoretical guarantee that smooth solutions to these equations can exist for all time. Mathematically, proving whether smooth solutions to these equations without external forces exist for all time or can break down in finite time has been a longstanding open problem. The potential breakdown mechanism also remains elusive for several related equations with a wide range of applications. The goal of this research is to investigate the potential breakdown mechanism for various equations and develop analytic and numeric tools that provide a theoretical understanding of these mechanisms. This award will also provide opportunities for students to be involved in the latest developments from this research through topics courses and research projects. The project aims to understand whether the incompressible 3D Euler equations and related equations could develop a finite time singularity from a smooth initial condition with finite energy. Our approach builds on PI's recent works on singularity formation in incompressible fluids and the self-similar method for finite time blowup, which consists of the following three steps. Firstly, we construct an approximate blowup profile, which can be obtained either analytically or numerically. Secondly, we impose suitable normalization conditions and prove the nonlinear stability of the approximate blowup profile. Thirdly, we choose a suitable perturbation to construct initial data with desired property and obtain finite time blowup using a rescaling relation. Additionally, the project seeks to develop a novel approach for the stability analysis in the self-similar variables that is robust enough to be applied to study a larger class of nonlinear partial differential equations. 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 · 2026-05

Sensory epithelia of the mammalian vestibular inner ear show strict zonal organization of neural encoding strategies: afferents from the peripheral epithelium use regular spike timing for rate encoding, while afferents from central epithelium have irregular spike timing and are better at temporal encoding. The different codes convey different head motion information; for otolith organs such as the utricle, vibration is better conveyed by central striolar zones and head tilt by peripheral extrastriolar zones. In addition to zonal organization, the sensory epithelia have two receptor cell types – conventional type II hair cells and unusual type I hair cells Type II hair cells, like other hair cells, transmit by quantal transmission of glutamate packets, while type I hair cells transmit to afferents by unique nonquantal mechanisms involving ions flowing through open channels on either side of the synaptic cleft. Here we use these findings to probe how type I and II HCs contribute to information processing on the vestibular epithelium. Do they, like other features, contribute to the zonal mechanisms that produce irregular and regular afferent streams? Two aims tackle these questions at different levels of complexity and resolution: individual (Aim 1) and population (Aim 2) levels. Our model preparation is the excised mouse utricle, which detects horizontal linear accelerations and head tilt, and its attached afferent nerve. We focus on “dimorphic” afferents, which contact both type I and type II HCs. In Aim 1 we ask how quantal vs. nonquantal information is integrated in vestibular afferent firing. This is essential to understanding vestibular afferent signals, given that most afferents are dimorphic. Preliminary data suggest that the two synaptic transmission modes process sensory information with different frequency tuning, linearity, gain, and temporal resolution. We will stimulate individual afferents and document key physiological and anatomical properties to build a computational “dimorphic synapse” model with regular and irregular versions. The modeling will test mechanistic understanding of how hair cells transmit quantal and nonquantal signals to afferent dendrites, how the information is integrated, and the functional significance of zonal differences. In Aim 2, we ask how head motions are signaled by populations of regular and irregular afferents, collectively and moment-by-moment, and how quantal and nonquantal transmission influence the spatial distribution of activity. To gather the population data, we stimulate groups of hair cells or afferents and correlate their simultaneous individual activities as represented by increases in intracellular calcium levels. The results yield a bird’s eye view of how signaling for a given stimulus varies between and within zones, afferent types and hair cell types. The analyzed data will be used as input to a multi-tier population model of how signals are encoded at hair cell, synaptic, and spike generation stages. This research is necessary to understand how we compensate for our own motions as we navigate the world, ensuring stable vision and balance.

NIH Research Projects · FY 2026 · 2026-04

RESEARCH ABSTRACT: HEPATITIS C VIRUS TRAFFICKING IN HEPATOCYTES Many viruses have evolved distinct pathways of transmitting their infection between cells. This viral spread involves canonical pathways that release unprotected virions from the infected cell (thus potentially exposing the virion to neutralizing antibodies) and/or pathways that shield the virion from the humoral response (cell-cell spread, syncytia formation, and the release of virions in extra-cellular vesicles). Viral egress pathways that evade the humoral response are especially important to understand given the potential importance of manipulating viral susceptibility to this critical arm of adaptive immunity. Most viruses of the Flaviviridae appear to use both mechanisms of viral egress. We and others previously characterized the pathway of canonical hepatitis C virus (HCV) release, while canonical dengue virus (DENV) secretion has also been well characterized. In contrast, noncanonical pathways of virion spread in the Flaviridae are poorly understood. Our lab has a longstanding interest in Flaviviridae entry and spread. In our prior funding period, we developed a multi-pronged analysis (host genetics, pharmacological inhibition and single particle imaging) of polarized hepatoma organoids to characterize HCV entry and egress. The work established the most complete, detailed pathway of HCV entry, in addition to extensive preliminary data on HCV cell-cell spread. We expanded these studies into the non-canonical spread of DENV, focusing on a process we discovered years ago: DENV-induced lipophagy, which is the selective autophagic mobilization of lipid droplet stores to stimulate lipid metabolism. This process has been associated with the non-canonical release of DENV into extra-cellular vesicles. We have identified a central cellular player in this process: the selective autophagy adaptor NBR1. In this R01 renewal application, we will focus on extending and expanding our studies on HCV and DENV non-canonical virion spread. In the case of HCV, we developed quantitative cell-cell spread assays and identified putative cellular secretion pathways and cargo adaptors that modulate cell-cell spread. We propose to characterize the significance of these pathways using imaging of our 3-dimensional polarized hepatoma spheroid system. Additionally, we will collaborate for an ultra-structural analysis of HCV-infected hepatocyte cell-cell junctions cryoelectron microscopy tomography (cryo-ET). For DENV non-canonical spread, we will validate a rove for NBR1 and lipophagy in the cell-cell spread of DENV. We will then characterize the pathway of non-canonical release of DENV virions. Finally, we will perform ultrastructural analysis of the NBR1-dependent DENV containing extra-cellular vesicles using cryo-ET. The specific aims are: Aim 1. Characterize the pathway(s) of HCV cell-cell spread in infected hepatocytes. Aim 2. Characterize the NBR1-dependent pathway of non-canonical DENV egress.

NIH Research Projects · FY 2026 · 2026-04

PROJECT SUMMARY Out-of-hospital cardiac arrest (OHCA) is a leading cause of death in the United States (US). Advances in resuscitation science have improved survival rates in some communities, but wide variability in incidence and survival outcomes persist. Multiple geographic and temporal factors contribute to this variability such as comorbidity burden, place effects, access to healthcare, poverty, community resources, and variation in clinical care policies. Given the substantial public health burden of OHCA and marked geographic variability in incidence and survival, developing a targeted framework to identify and measure OHCA incident and outcome risks is essential. We will employ a participatory and mixed methods approach that combines machine learning (ML) and artificial intelligence with qualitative research methods to develop and evaluate an OHCA risk score and a virtual laboratory (VL) environment as decision support tools to inform community-level interventions to improve OHCA outcomes. We will first engage community representatives and officials, involved in the OHCA system of care (e.g. community service organizations, emergency medical service providers, hospital quality assurance officers, public health officials, and cardiac arrest survivors) to participate in focus groups and key informant interviews to identify optimal and efficient data elements to define a scalable and usable OHCA risk score (Aim 1). Based on this information, we will then employ ML methods to develop the OHCA risk score and VL environment (Aim 2), which will then be discussed and evaluated by community representatives (Aim 3). These elements of participatory ML will provide important context for data interpretation while building trust in the OHCA risk score and VL environment as pre-implementation tools to diagnose local delivery capabilities and develop implementation strategies to overcome any barriers identified. The OHCA risk score and VL environments resulting from this project can inform public health messaging, aid local public health departments and hospitals to identify areas where surveillances needs to be heightened, and inform government agencies where to direct funding and resource allocation as it pertains both to the chain of survival as well as prevention and early identification of patients at risk for OHCA. This work is a necessary first step to direct strategic investments in emergency response infrastructure and community-level interventions to improve preparedness and optimize OHCA survival outcomes.

NIH Research Projects · FY 2026 · 2026-04

PROJECT SUMMARY Our long-term goal is to understand how embryos evolve. Many developmental mechanisms are similar between species as different as humans and flies, but occasionally they differ strikingly even between closely related species. Explaining such unexpected plasticity in developmental gene networks will help us understand the basic science of developmental robustness and congenital disease in all animals, including humans. Comparing multiple closely related species is a powerful approach for understanding causes of plasticity in gene networks but is extremely difficult to implement with vertebrates. However, it can be accomplished with flies (Diptera) because many fly species are readily cultured at low cost and are amenable to functional studies. Moreover, a leading genetic model organisms in developmental biology, the fruit fly Drosophila melanogaster, can serve as entry point for comparative functional studies. We therefore established experimental tools and genomic resources for a set of non-traditional dipteran model organisms to study how developmental mechanisms change and diversify in the course of evolution. We discovered that different fly species use unrelated anterior determinants (ADs) to establish the anterior-posterior (AP) axis of the embryo. While Drosophila’s AD (Bicoid) is a widely studied morphogen model, lower dipterans use a range of unrelated transcription factor encoding AD genes. Our recent research shows that ADs of two lower dipteran species, the moth fly Clogmia albipunctata and the harlequin fly Chironomus riparius, differ in their target genes and mechanisms of action (e.g., activation versus repression). This discovery was possible after generating genomic resources for our model organisms and combining knowledge of DNA-binding motifs of the disparate ADs with quantitative measurements of chromatin accessibility and gene expression in precisely staged single embryos with normal or reduced AD activity. We also obtained preliminary evidence that the mechanisms for initiating axial symmetry-breaking in higher dipterans (Brachycera), such as the soldier fly Hermetia illucens, a close outgroup to the clade of species in which bicoid occurs, or the flesh fly Sarcophaga bullata which lost bicoid, differ more radically. It is not known to what extent the segmentation gene networks differ between dipterans and how they converge during embryogenesis to establish the conserved segmented body plan. We will use these and additional species as entry points to examine through the expanded use of scalable (epi-)genomic approaches and functional experiments in vivo (1) how dipterans diversified their segmentation gene networks in evolution while converging in development on a segmented body plan and (2) how and why molecular mechanisms of axial symmetry breaking diverged. Answers to these questions will help to explain how, when, and why developmental gene networks of embryos reorganize.

NIH Research Projects · FY 2026 · 2026-04

Project Summary Adolescence - the developmental stage between childhood and adulthood - is associated with robust neural and behavioral changes, including greater connectivity between brain regions that are involved in decision-making functions and increased sensitivity to social rewards. Social experiences that occur during adolescence can have a lasting effect on the individual, as adolescent social isolation is associated with impaired decision-making functions and greater drug use in adulthood. We, and others, hypothesize that isolation during adolescence disrupts the neurodevelopmental processes that regulate addiction susceptibility. The identity of these mechanisms, however, is not known. This proposal will determine how social isolation during key developmental periods disrupts gene regulatory mechanisms crucial for the development of orbitofrontal circuits that regulate decision-making and addiction susceptibility. To accomplish this goal, this proposal will integrate sophisticated behavioral assays with computational tools, drug self-administration, and single-cell genomic analyses across adolescence and adulthood in the rat to identify the neurodevelopmental mechanisms that are altered by adolescent social isolation and associated with greater risk for drug use and addiction-like behaviors. In Aim 1 we will identify the developmental window in which decision-making functions are most sensitive to social experiences by assessing decision-making trajectories in rats that are socially isolated or housed at different adolescent ages. We will then conduct the first in-depth characterization of cocaine-taking behaviors in Aim 2 to identify the behavioral mechanism(s) responsible for the elevations in cocaine use observed in adult rats who were isolated during adolescence. Finally, in Aim 3 we will identify the gene regulatory mechanisms within the amygdala-orbitofrontal circuit that are responsible for social isolation induced decision-making impairments using circuit-specific single-nucleus RNA-sequencing and single-nucleus DNA methylation sequencing. These innovative and integrative series of studies will generate a novel framework linking circuity-specific gene regulation with complex behaviors to understand how adolescent social experiences impacts the precise neurodevelopmental mechanisms that mediate addiction susceptibility.

NIH Research Projects · FY 2026 · 2026-04

ABSTRACT Excessive alcohol use contributes to the deaths of over 100,000 Americans and costs the United States over $249 billion each year. Understanding factors that contribute to the development and persistence of excessive drinking is crucial for informing effective prevention, education, and intervention approaches. To date, the Chicago Social Drinking Project (CSDP) has contributed key insights into the role of subjective and physiologic responses to alcohol in the escalation and maintenance of excessive drinking and alcohol use disorder (AUD). Our unique integration of human laboratory alcohol challenge and longitudinal follow-up of drinking showed that young adult (ages 21-35 yrs) heavy social drinkers (HD) at risk for AUD show heightened sensitivity to alcohol stimulation and reward (liking, wanting) versus light drinkers (LD), and that these responses (versus low sensitivity to alcohol sedation) were more predictive of future drinking escalation and development of AUD in the transition to middle age and beyond. Importantly, this positive alcohol response phenotype was robust and reproducible in a second HD cohort. As a result, the CSDP has challenged conventional notions of vulnerability to AUD by showing that sustained and heightened sensitivity to alcohol stimulation and reward, rather than low level responses, are primary predictors of drinking escalations over time. Moreover, positive alcohol responses were maintained or magnified across a decade of re-examination testing in drinkers progressing on AUD symptom count over time. This proposed CSDP renewal award is timely, as we will expand our program of research across the lifespan and enroll very young adult drinkers (age 18-19), extend follow-up to middle-age in the second cohort, and conduct analyses examining alcohol responses in older chronic AUD and age-matched LD. In Aim 1, high-resolution ecological momentary assessment (HR-EMA) will be employed to provide the first real-time testing of alcohol sensitivity in 18-19-year-old high- and low-risk very young adult drinkers. This ambulatory assessment circumvents restrictions on providing alcohol to persons <21 yrs. We will also determine the prospective role of alcohol response phenotype (sensitivity vs. insensitivity) in predicting their future alcohol use and problems over two-year follow-up. In Aim 2, we will conduct a final 15- yr follow-up in our second cohort of HD (enrolled 2009-2011, now entering their 40s) to serve as a reproducibility sample and enhance robustness of our long-term predictive models of AUD risk. In Aim 3, we will conduct analyses of natural environment alcohol responses in our cohort of older adults with chronic AUD (ages 40-65 yrs, >20-year AUD duration) and age-matched LD to investigate if older AUD drinkers show sustained positive alcohol responses like young adult AUD drinkers, or the purported allostatic shift to reduced alcohol reward and drinking to relieve negative affect. The proposed work will further our understanding of subjective alcohol responses and drinking behaviors in individuals from late adolescence to older adulthood, enhancing our understanding of this critical factor in the vulnerability, severity, and maintenance of AUD.