Yale University

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY / ABSTRACT Alcohol use disorder (AUD) is a chronic debilitating condition that accounts for over half of all substance abuse treatment cases in the United States. Most AUD patients relapse despite treatment. It is increasingly recognized that deficits in avoidance learning are critically involved in motivating habitual and heavy alcohol use. Specifically, due to alcohol’s anxiolytic and analgesic properties, many engage in drinking to avoid painful physical and affective states. Paradoxically, chronic alcohol use is associated with increased pain reactivity and decreased cognitive control. These changes reinforce compulsive drinking as a maladaptive avoidance strategy and further compromise learning of its harmful consequences. Yet, the neural, physiological, and psychological processes inter-relating avoidance learning dysfunction with the maintenance and relapse of AUD remain poorly understood. This K99/R00 proposal addresses this critical gap in research. To this end, we propose to collect functional magnetic resonance imaging (fMRI) data in treatment- seeking AUD patients during a probabilistic learning task which features pain and reward. Our first aim is to characterize dysfunctions of the brain circuits supporting pain reactivity and cognitive control during avoidance learning in AUD patients, as compared to social drinkers. In the second aim, we will evaluate how the neural markers, along with clinical, physiological and behavioral metrics, may be used to (1) diagnostically distinguish clinical characteristics; and (2) model the key pathophysiological pathways that sustain habitual alcohol use. In the third aim, we will identify the risk factors that best predict relapse during the 12-month follow-up, thus offering clinical implications for improving treatment outcomes. The long-term goal of the candidate is to start an independent career in neuroscience research of alcohol addiction. This proposed study will support this goal by serving as a launchpad for the candidate to transition to an independent investigator. The candidate has trained extensively in cognitive neuroscience and devoted himself to the field of addiction neuroscience. By conducting this study, the candidate will broaden his training in the clinical and neurobiological investigation of AUD as well as gain expertise in machine learning and Bayesian modeling. The candidate has identified his training needs and assembled a team of expert mentors for this K99/R00 proposal. The training plan includes structured mentoring, supervised research, formal coursework, presentations at scientific meetings, and professional development. The exceptional research environment and intellectual resources at Yale University will allow the candidate to receive ample guidance, learn novel techniques, and gain independence, while pursuing the research he is passionate about.

NIH Research Projects · FY 2025 · 2024-09

ABSTRACT Cellular behaviors are modulated by a variety of stimuli in the environment, which can be generally categorized as chemical, physical, and biological factors. Previous work suggests that physical forces, both at the mesoscale of cell-cell interactions and the microscale of ligand-receptor binding events, play a critical role in regulating cell physiology. For example, stem-cell differentiation is regulated by neighboring cells and the stiffness of the extracellular matrix, while T cells are better activated by target cells with a mechanically strong cortex. The T- cell receptor also bears forces exerted from the pMHC, which affects receptor activation in a non-monotonic manner. While a plethora of evidence suggests the critical role of mechanical force in regulating cell fate and activation states, force itself has rarely been considered or exploited as a target for cell engineering and therapeutics development. Major hurdles include the lack of effective sensors that can digitally sense mechanical forces, and the lack of genetically encoded intracellular devices that can convert mechanical forces into a signaling cascade that modulates cell states. In this proposal, we will combine state-of-the-art technologies in protein design, RNA synthetic biology, and cell engineering to develop “mechaswitches”: universal, modular and programmable signal transduction systems that are able to trigger specific cellular actions in response to mechanical signals. Each mechaswitch is composed of a protein-based force sensor inserted within a force- bearing protein of interest and a transducer mRNA that implements a desired cellular action. When the protein of interest is subjected to a defined range of forces, the mechaswitch force sensor responds by changing its conformation. This conformational change is in turn detected by a sensing element within the transducer mRNA that either switches on or off the expression of specific proteins that modulate cell behavior. We will develop sensors covering a wide range of molecular forces and RNA switches that control protein expression with high fidelity. In proof-of-principle experiments, we will apply mechaswitches to program cell differentiation in defined extracellular environments and to modulate T cell proliferation and activation in response to specific antigens. Because of the modularity of mechaswitches, which enables their sensing and output components to be rapidly swapped and recombined, we envision a host of other uses for mechaswitches that could transform the study and application of mechanical forces in cell biology. The successful realization of this project is expected to not only advance basic research in mechanobiology, but also lay the foundation for the first therapeutic strategies informed by the mechanical signals and target mechano-properties of diseased cells and proteins.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY/ABSTRACT Gastric ulcer is increasing in the United States due to increasing use of non-steroidal anti-inflammatory drugs (NSAIDs) and Helicobacter pylori infection. It has been a major cause of gastrointestinal surgery with high morbidity and mortality. While molecular and cellular mechanisms associated with gastric ulcer repair have been heavily investigated, information on the cell of origin for regenerated epithelium is limited. Microscopically, gastric ulcer is characterized by mucosal injuries involving the muscularis mucosae and a cavity surrounded by acute and chronic inflammation. Numbers of acid secreting parietal cells are decreased in the ulcer edges. Parietal cells are also decreased in preneoplastic lesions such as atrophic gastritis and Ménétrier’s disease. We have previously demonstrated that parietal cells undergo apoptosis in an atrophic gastritis mouse model. Ménétrier’s disease is a rare, acquired, premalignant protein-losing hypertrophic gastropathy induced by EGF receptor (EGFR) signaling activation. Ménétrier’s disease also shows decreased number of parietal cells, however, apoptosis is not changed. This raises the possibility that parietal cells transdifferentiate into other cell types rather than dying in Ménétrier’s disease. As a part of my K08 project, we examined the fate of parietal cells in a Ménétrier’s disease mouse model and observed that parietal cells can give rise to other epithelial cell types. Our preliminary data shows that parietal cells can transdifferentiate into various other epithelial cell types in another EGFR activated condition gastric ulcer. Preliminary analysis on single cell RNA sequencing (scRNA-seq) comparing normal stomach and gastric ulcer at day 3 showed upregulated Sox4 expression and activated FoxO signaling pathway in parietal cells from gastric ulcer. Sox4 and FoxO signaling have been associated with stem cell maintenance, transdifferentiation, and cancer progression. It has been reported that spasmolytic polypeptide expressing metaplasia (SPEM) develops at the ulcer edges and plays an important role in gastric ulcer healing. We will investigate if parietal cells can transdifferentiate into SPEM in gastric ulcer and give rise to other epithelial cell types during gastric ulcer healing by lineage tracing followed by immunofluorescence microscopy and single cell RNA sequencing. The role of Sox4 in parietal cell plasticity will also be investigated by knocking out or overexpressing Sox4 in parietal cells. Successful completion of the studies proposed in this application will not only provide us with better understanding of the underlying mechanism of gastric epithelial plasticity but also with a foundation for effective treatment of gastric ulcer. Also, comparing results from this project with my K08 project will contribute to my long- term research goal of preventing gastric tumorigenesis and facilitating gastric mucosal regeneration.

NIH Research Projects · FY 2024 · 2024-09

Problem: There is high unmet need for PrEP among women in the U.S. Even where resources are available to support PrEP, it remains inaccessible to women because too few clinical settings provide it. Purpose: To support women’s choice, we will expand and enhance a PrEP decision aid to include the full array of available formulation options and then integrate its delivery into sexual and reproductive health settings during counseling visits, where PrEP can be provided to all women who are interested. Methods: Project Carmenta is guided by health-focused implementation frameworks, critical ongoing input from our Community Advisory Board and the cumulative expertise of the study team in HIV prevention, sexual and reproductive health, decision science, implementation science, and epidemiology. We will engage patients (n=15-20), clinicians (n=10), and staff (n=10) at sexual and reproductive health clinics across Greater New Haven, Connecticut for semi-structured interviews to expand and enhance an existing PrEP decision aid to include all available formulations and optimize its integration into sexual and reproductive health clinics. Qualitative interviews will inform infrastructure development to support PrEP delivery in sexual and reproductive health clinics. In a hybrid Type 2 effectiveness-implementation study, we will then randomize patients (n=50) to receive either the PrEP decision aid or generic PrEP information prior to a clinician visit. In follow-up interviews immediately post-visit, and at Months 3 and 6, primary outcomes are clinical efficacy (PrEP initiation) and implementation (using Proctor definitions for feasibility, acceptability, penetration, and adoption) that are important for future planned scale-up. Aims: The Aims of the project are: 1) To expand and enhance an existing decision aid on PrEP, and assess innovation, clinical encounter, recipients, and context- level factors affecting its integration into counseling visits in sexual and reproductive health clinics; and 2) To integrate the expanded patient-facing individual PrEP decision aid into counseling visits and evaluate clinical and implementation outcomes. Significance: The proposed research directly addresses the key objectives and priorities of the NIH Office of AIDS Research by using a decision aid to support women’s choice of PrEP product and providing new insights on how women weigh each option. Significance is high because integration of PrEP into sexual and reproductive health reduces barriers to access for women. Innovation is high because of the use of a trauma-responsive PrEP decision aid for women within a sexual and reproductive health setting. Public health impact is high because of a focus on supporting health in a rapidly changing sexual and reproductive health landscape and supporting women’s choice of PrEP.

NIH Research Projects · FY 2025 · 2024-09

Scalable interventions that facilitate adherence to HIV pre-exposure prophylaxis (PrEP) are urgently needed to address the alarming HIV incidence in in the U.S. PrEP reduces the risk of acquiring HIV by >92%; but it has not achieved implementation at a scale sufficient to curb the growth of the US HIV epidemic. There are multi-level barriers to PrEP adherence, including interpersonal and structural barriers. To optimize health in PrEP coverage, individuals must have access to service options that reduce their exposure to these barriers. Many Men Many Voices (3MV) is a group-level behavioral intervention that demonstrated efficacy for increasing healthcare-seeking behaviors. Client-centered care coordination (C4) addresses social barriers through a service model that trains staff to deliver autonomy-supportive care and addresses structural barriers via an integrated online platform to improve the continuity of coordination between service providers. Our goal is to combine two evidence-based interventions into a multi-level integrated strategy that directly addresses interpersonal barriers via 3MV, extends 3MV effects into service-delivery settings by training staff on key concepts, and then addresses structural barriers by linking men to a network of services via the online C4 platform. We will pursue the following aims with 2 implementing agencies located in high HIV incidence communities (1) determine the efficacy of an integrated 3MV+ C4 for increasing PrEP adherence; (2) ascertain the optimal dose of C4 implementation for maximizing its effect on PrEP adherence; (3) identify the critical leverage points in the implementation ecosystem to target for change. 48 egocentric networks of men (N=480; mean network size n=10) will be recruited in two collaborating sites in Dallas/Ft. Worth and Bronx/Harlem. Networks will all begin with a 3-month control phase and be randomized into either the C4 arm or the C4+3MV arm. The Learn-as-you-go (LAGO) statistical methodology will be used to optimize the strategy and associated costs by generating statistical recommendations for modifying pre-specified components of C4. Qualitative interviews will explore the factors affecting variation in implementation between the three cities. Lastly, we will develop a policy brief accounting for the political economy and non-economic welfare effects of 3MV+C4. This study advances HIV prevention science by generating evidence for an intervention that will contribute to health optimization in the impact of PrEP on the HIV epidemic. Our research also uses innovative statistical and interdisciplinary methods that allow us to maximize 3MV and C4’s effects on adherence by: (1) optimizing elements observed to be contributing to increases in adherence and (2) streamlining excess elements to reduce implementation time and costs; thus, enhancing scale-up potential in End the HIV Epidemic communities.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY Among the ~2.5 million persons who develop incident tuberculosis (TB) in sub-Saharan Africa annually, about one in five does not complete treatment, even though highly effective treatments for TB and concurrent HIV are now widely available. This contributes significantly to the high TB mortality worldwide, especially among persons living with HIV. Clinical and public health guidelines universally recommend TB education and counseling (TB- EC) to improve adherence to and outcomes of TB treatment and antiretroviral therapy (ART), but many barriers to delivery exist. There is growing community interest in adapting peer-led strategies from the HIV field to improve provision of TB-EC and outcomes of TB and HIV. We recently developed and implemented a multi-component peer-navigation strategy for delivery of TB-EC among persons with TB with and without HIV in Uganda, including (1) task-shifting of TB-EC to peers with TB; (2) restructuring of clinic workflows; (3) a TB-EC checklist; (4) individualized adherence planning; and (5) behavior-change messaging. In a preliminary evaluation, we found that the adapted TB-EC strategy was feasible and acceptable, and improved both TB literacy and treatment outcomes. We now propose a 16-site, cluster-randomized, hybrid Type 2 effectiveness-implementation trial to evaluate the impact of peer-navigation on TB/TB-HIV treatment adherence and clinical outcomes among new PWTB with and without HIV in Uganda. We will conduct these studies through the Uganda TB Implementation Research Consortium (U-TIRC), an academic-public health partnership hosted at the Walimu non-governmental organization and involving Makerere University, the Uganda National TB Program (NTP), New York University, and Yale University. Our primary effectiveness outcomes include TB treatment completion and ART retention at one year. Our primary implementation outcomes include TB treatment and ART initiation, adherence, and persistence. We will conduct rigorous mediation analyses to test our hypotheses about the social and behavioral mechanisms through which peer navigation strategy was designed to influence client outcomes. We will nest convergent mixed-methods studies of implementation fidelity and context to understand provider and health system influences. Our overall hypothesis is that peer navigation will improve TB/TB-HIV treatment adherence and clinical outcomes compared to standard TB-EC by addressing individual and health-system barriers to TB treatment and ART. These studies address several NIH research priorities, including overcoming barriers to implementing evidence-based interventions for TB-HIV and improving medication adherence. We expect that our rigorous design and implementation plans will provide high-quality data on the effectiveness and implementation of a novel peer-navigation strategy to promote TB treatment and ART adherence. We will disseminate our findings to participants, the Uganda National TB Program, and the global TB and HIV communities to inform how TB-EC is offered to persons with TB/TB-HIV. If successful, our study will lead to future work evaluating the impact, sustainability, and cost-effectiveness of the peer-navigation strategy at scale.

NIH Research Projects · FY 2026 · 2024-09

ABSTRACT The Community Partnerships to Advance Science for Society (ComPASS) Health Equity Research Hub at Yale (Yale ComPASS Hub) will support select community-led health equity structural interventions (CHESIs) as identified by ComPASS leadership and will synchronize with the ComPASS Coordinating Center and the broader ComPASS program to provide thematic technical assistance across food/nutrition security, intergenerational family interventions, poverty-reduction, cultural resilience, and participatory research and implementation approaches. These thematic areas represent the extensive expertise of the team and address challenges encountered in diverse settings throughout the U.S. The Yale ComPASS Hub has the capacity to expand the breath of technical assistance thematic areas as CHESI needs are identified. The Yale ComPASS Hub will use community-academic partnerships to tailor and deliver responsive, accessible, and transformational technical assistance to CHESIs as they generate new scientific knowledge about community- led structural interventions to advance health equity. The overall aim of the Yale ComPASS Hub is to provide evidence-derived technical assistance and capacity-building support to CHESIs based on collaborative processes and practical guidance adapted by partnered academic and community leader Hub unit teams. Hub units will each be co-led by Academic and Community Director teams with almost two decades of experience working collaboratively on health equity-related research or practice initiatives. The Yale ComPASS Hub will elevate the role of partnered research and decision-making in grant governance and implementation, bring diverse methodological and analytical research expertise to facilitate transdisciplinary collaboration, meaningfully engage community members to more equitably partner in co-creation and implementation of research, and elevate the science of community engaged research to drive health equity.

NIH Research Projects · FY 2024 · 2024-09

PROJECT SUMMARY/ABSTRACT Irritability, an increased propensity to experience anger and frustration, is among the most common presenting complaints in child psychiatry. Severe irritability affects up to 10% of youth in the U.S. and causes significant impairment and high rates of service use and school suspensions. Normative irritability follows a low, declining trajectory from childhood to adolescence. However, some youth remain persistently, highly irritable over development, putting them at the greatest risk for later psychopathology and adverse outcomes, including depression, anxiety, and suicidality. Currently, there are no evidence-based treatments for chronic irritability. This is because of the limited understanding of the etiology and mechanisms of chronic irritability. This New Innovator application addresses this significant gap by identifying predictors and mechanisms of chronic irritability trajectory across multiple levels of analysis including the brain, physiology, behavior, social experience, and family/environment. Specific goals of this project are to (1) identify neural markers (i.e., brain function and connectivity) that predict which child will follow a chronic, persistently high irritability trajectory over time and the developmental changes in these neural markers that underlie chronic irritability; (2) identify social and environmental determinants of chronic irritability and their mediating effects on the link between neural alterations and chronic irritability; (3) enhance prediction of chronic irritability using data from multiple levels of analysis including the brain, physiology, behaviors, social experiences, and familial/environmental factors. We will accomplish these goals in a 3-wave longitudinal functional magnetic resonance imaging (fMRI) study, with a sample of 180 children with (n=120) and without elevated irritability (n=60) at ages 8–13 years. Multi-method, multi-informant assessments will occur annually at three timepoints over two years. Parent-, youth-, and clinician- reports will assess youth irritability. An innovative smartphone-based, naturalistic ecological momentary assessment will measure irritability as well as parent-child and peer relationships/interactions in real time. fMRI will be collected during a novel social vs. non-social FNR task, addressing a significant gap in the field by probing the role of contexts in frustration-related neural alterations. Parent and youth will engage in two interactive tasks in the laboratory (conflict discussion and solving difficult puzzles) while their behaviors and heart rates are being measured. We will first chart within-person irritability trajectories over time using data-driven, person-centered latent class growth mixture modeling, and then apply machine learning with the multi-level data to predict chronic irritability trajectory at the single-subject level. This project, a significant leap forward from the current field, will provide novel, important insights into the predictors and mechanisms of irritability trajectories across multiple levels of analysis. Results will advance efforts toward development of evidence-based preventions and interventions for irritability—a top problem in child psychiatry and a robust predictor of suicidality.

NIH Research Projects · FY 2025 · 2024-09

Project Summary Heart failure is a leading cause of morbidity and mortality, projected to affect more than eight million Americans by 2030. Frequent symptoms, hospitalizations, and hospital readmissions are common among this population, causing significant suffering, particularly near the end of life. Conversations about prognosis, the benefits and harms of invasive treatments, and patient care goals and preferences are essential to promote goal- concordant care among this population. Yet, these conversations occur infrequently. To address these critical needs, the American Heart Association, the American College of Cardiology, and the Heart Failure Society of America recommend palliative care as a Class I (strong) recommendation “for all patients with heart failure to improve quality of life and relieve suffering.” However, fewer than 20% of this population receive Specialist Palliative Care. Patient prognosis is a key criterion for referral to palliative care among this population. Our team developed and validated an automated electronic health record-based 1-year mortality risk model for people hospitalized with heart failure and embedded the model into an interruptive alert to facilitate clinician decision-making. While the alert was highly accurate, acceptable, and feasible among clinicians, it did not result in a significant increase in referrals to Specialist Palliative Care in a subsequent clinical trial. Interruptive alerts deliver the right information but often at the wrong time, resulting in clinician overrides and task abandonment. Non-interruptive clinical decision support (CDS), such as targeted modal alerts and passive designs, provides innovative alternatives to interruptive alerts. However, their effectiveness hinges on careful user-centered design and the development of strategies for implementation. Thus, our overall objective is to design and pilot test the Promoting Palliative Care for People with Heart Failure (P3HF) CDS, a tool that incorporates our 1-year mortality alert with evidence-based CDS design, and critically, pairs the CDS with strategies for implementation. To accomplish our objectives, we will design the P3HF CDS and develop an accompanying implementation package using a theory-informed process involving interviews, iterative design, and focus-group-led evaluation with healthcare provider stakeholders (Aim 1). Next, we will conduct a randomized real-world pilot trial across two hospitals to evaluate the usability, acceptability, appropriateness, and feasibility of the P3HF CDS and implementation package. We will also examine the intervention's preliminary efficacy on clinician behavior change (i.e., referral to Specialist Palliative Care, primary outcome) and patient outcomes (i.e., Advance Care Plan documentation, hospice enrollment, hospital length of stay, 30- day hospital readmission, exploratory secondary outcomes; Aim 2). This study will provide essential data necessary and sufficient to inform a subsequent multi-site hybrid efficacy-effectiveness clinical trial of the P3HF CDS and implementation package. If successful, this project will result in an evidence-based, generalizable intervention that could greatly improve care delivery and quality of life for people with heart failure.

NIH Research Projects · FY 2025 · 2024-09

Non-Hodgkin lymphomas (NHL) represent about 90% of all lymphomas diagnosed each year and is classified based on cell type - B cell, T cell and natural killer (NK) cell types, location - nodal or extra nodal, and the tumor grade - aggressive (high grade) and indolent (low grade). Follicular lymphoma (FL) is the most common indolent B-cell lymphoma but remains a largely incurable malignancy. The most clinically challenging aspect of FL is the transformation into diffuse large B-cell lymphomas (DLBCL), characterized by the emergence of more aggressive subclones, loss of the follicular growth spatial architecture, and resistance to treatment, leading to a much shortened survival period, typically less than 2 years. Among T-cell lymphomas, angioimmunoblastic T-cell lymphoma (AITL) is one of the most common subtypes characterized by a tumor with follicular helper phenotype surrounded by an inflammatory microenvironment, arborizing vasculature, and progression with dramatic changes in spatial architecture. Although the discovery of FL transformation and AITL tumor evolution was initially documented over 50 years ago, the biological mechanisms and clinical implications remain poorly understood. No biomarkers exist to predict or therapies to prevent its metastases or progression to highly aggressive lymphomas. Both of these tumors hijack normal follicle biology to escape immune surveillance and potentially develop resistance clones. Yale HTAN Center aims to leverage the latest development in single-cell and spatial omics technologies to construct a spatiotemporal atlas of human FL transformation to DLBCL and AITL evolution. Specifically, we will apply high-plex immunofluorescence protein imaging to map all major cell types and spatial whole transcriptome sequencing to link cell type to mutational landscape, clonal evolution, and spatial interaction within the tumor microenvironment. We will further integrate spatial omics data with single nucleus RNA sequencing to identify cell subtypes and niches across tissue samples over various disease stages to construct a complete cell atlas associated with tumor transformation for different sexes and racial/ethnic groups and then computationally model the spatiotemporal evolutionary dynamics. Finally, we will apply and integrate spatial-epigenome-transcriptome co-profiling to unveil epigenetic mechanism underlying such transformation and potentially discover earliest events to predict the progression. The proposed spatiotemporal lymphoma atlas represents a valuable resource to test a range of hypotheses such as how different tumor clone emerge, interact, compete or cooperate in the spatial tissue context to drive lymphomagenesis, how T cells recognize and interact with different mutant clones, how the microenvironment co-evolves with tumor cells, and how to predict the likelihood of transformation and therapeutic stratification of patients. Single-cell spatial omics techniques and computational models can be applied to other types of human tumors within the HTAN consortium.

NIH Research Projects · FY 2025 · 2024-09

PROJECT ABSTRACT E-cigarettes are the most popular nicotine-delivery devices used by US adolescents. Many adolescents do not like being addicted to e-cigarettes and want to quit. However, there are no empirically validated interventions to help them quit using e-cigarettes. We have developed an adolescent-focused smartphone-app called Kick-Nic!© that uses an engaging, multi-media rich, and interactive cognitive-behavioral therapy platform to teach coping skills for e-cigarette-specific triggers and support cessation. App content to address e-cigarette appeal, quit motivations, and coping skills, was developed using focus group evidence from 60 adolescent e-cigarette users. The structure, format and design of the app was developed using iterative feedback from 14 adolescents. A feasibility pilot with 19 adolescents who used the app, indicated that it was engaging, easy to use, and useful to support e-cigarette quit efforts. An open label pilot is testing the use of the app with engagement methods commonly used in digital interventions that were identified by youth as important for engagement [text message reminders & weekly in-person check-ins]; 8 high school adolescents have initiated the trial, are completing all app sessions, weekly check-ins, and assessments with 100% retention. The current proposal will: Aim 1: Examine the efficacy of the Kick-Nic!© App: We will conduct an RCT with 306 adolescents in high schools who use e-cigarettes regularly (>/=1 day/week in the past month) and want to quit. Adolescents will be randomized to the ACTIVE (Kick-Nic!© App for eight weeks, along with text message reminders and weekly virtual check-ins, & assessments) or the CONTROL condition (assessments alone with referrals to NCI Quit Vaping webpage). E-cigarette use will be assessed at baseline, biweekly during treatment, and end of treatment (EOT; 8 weeks), and then at 1, 2, 3 and 6 month follow ups (FU). Salivary cotinine levels </= 30 ng/ml) will verify self-reports of abstinence at EOT and 6-month FU. We will explore changes in other tobacco and cannabis use behaviors. Our primary outcome will be 7-day, biochemically verified, point-prevalence abstinence (PPA) rates at 6-months, and secondary outcomes include 7-day PPA at EOT, % days e-cig free (during treatment, at EOT, 3 and 6 mths FU) and continuous abstinence (at EOT, 3 and 6 mths FU). We will also examine the impact of app engagement, coping skill knowledge, and baseline variables (nicotine dependence, e-cig use frequency, sex) and other tobacco product use on outcomes. Aim 2: Identify strategies for dissemination and implementation of the app within schools. Individual qualitative interviews with up to 50% of youth in the ACTIVE condition will examine likeability of the app and engagement methods used, how to disseminate the app to youth in school settings. Interviews with 20 school staff/administrators will examine how the app could be implemented and used within the school setting. Thus, this innovative proposal will support the testing, as well as dissemination and implementation within schools, of an appealing digital app that addresses the critical need for an adolescent-focused e-cigarette cessation intervention.

NIH Research Projects · FY 2025 · 2024-09

SUMMARY Cholestatic liver diseases such as primary sclerosing cholangitis (PSC) and primary biliary cholangitis (PBC) are autoimmune liver diseases with limited treatment options. Lymphatic vessels, located in the vicinity of bile ducts, help to reduce inflammation by removing cellular waste materials and inflammatory immune cells from the liver. Further, lymphatic endothelial cells (ECs) likely have contact with bile acids, as bile acids and bilirubin have been observed in lymph of animals with cholestasis. The role of the hepatic lymphatic system in cholestatic liver disease is largely unknown, and so are the effects of bile acids on lymphatic EC functions. The goal of this proposed study is to explore the potential of increasing lymphatic vessels as a therapeutic option for cholestatic liver diseases and understand this healing process with a focus on the microenvironment and the molecular mechanisms underlying lymphangiogenesis (formation of new lymphatic vessels). Our preliminary studies showed that lymphatic vessel numbers were significantly increased in mouse models of cholestatic liver disease. We have also found that promoting lymphangiogenesis through overexpression of vascular endothelial growth factor C (VEGF-C), the most potent lymphangiogenic factor, can significantly decrease liver fibrosis and injury in mice with cholestasis, suggesting a beneficial role of lymphatic vessel formation in cholestatic livers. Further, we have determined that liver lymphatic ECs express bile acid receptors, such as Takeda G protein-coupled receptor 5 (TGR5), and significantly proliferate in response to bile acids such as taurolithocholic acid (TLCA) and taurodeoxycholic acid (TDCA), known to have a strong affinity to TGR5. These observations have led us to hypothesize that hepatic lymphatic vessels may exert a beneficial effect on cholestatic liver injury and that certain hydrophobic bile acids may be novel pro- lymphangiogenic factors. To test these hypotheses, we propose three Aims: 1) determine the role of the hepatic lymphatic system in cholestatic liver disease, 2) determine the mechanism of bile acid-driven hepatic lymphangiogenesis, and 3) determine the spatial and temporal regulation of lymphatic vessels and their impact on the progression of cholestatic liver disease. The proposed study will determine whether hepatic lymphatic vessels work to ameliorate cholestatic liver disease and whether boosting lymphangiogenesis and lymphatic drainage could be a new therapeutic option for this disease.

NIH Research Projects · FY 2025 · 2024-09

Motile cilia beat rhythmically to propel cell movement or drive extracellular fluid flow. The functional importance of cilia motility in human health is highlighted by primary ciliary dyskinesia (PCD), a genetic disease caused by cilia motility defects. Patients with PCD display left-right asymmetry defect, reduced fertility and progressive lung disease. Currently there is no specific therapy for PCD and management of symptoms has been the main approach. The dynein arms that power cilia motility comprise multiple components that are pre-assembled in the cytosol and many genes associated with PCD encode components of dynein arms. A separate group encode dynein arm assembly factors (DNAAFs), proteins that reside in the cytosol and facilitate the assembly of dynein arm subunits. Interestingly, multiple DNAAFs are localized in droplet shaped cytosolic foci. However, the precise function of these foci and the precise molecular function of most DNAAFs remain poorly understood. Based on extensive preliminary and published data, our central hypothesis is that the co-chaperone proteins Ruvbl1 and Ruvbl2 are core components of a novel membrane-less cytosolic assemblage that functions to coordinate the translation, folding and assembly of axonemal dynein arm components. In this project, we will combine zebrafish genetics, mouse genetics and cultured tracheal cells to test our central hypothesis. We propose two specific aims to achieve this goal. In the first aim, we will dissect the mechanism of Ruvbl1-Ruvbl2 foci formation. In the second aim, we will systematically define R2HAD components and dissect their biochemical and functional relationships with DNAAFs associated with PCD. Successful completion of this project will not only provide a molecular framework for dynein arm assembly and the etiology of PCD, but also lay the foundation for future investigation into the regulation, and possible intervention, of dynein arm assembly and cilia motility under diverse physiological and disease conditions.

NSF Awards · FY 2024 · 2024-09

Streams and rivers are natural sources of CO2 to the atmosphere. Many rivers have higher concentrations of CO2 than the atmosphere, which creates a movement of CO2 from the river to the air. Wetlands and floodplains can contribute large amounts of this greenhouse gas to rivers and streams when they are connected by overland flooding. The goal of this project is to determine what types of wetlands and floodplains add the most CO2 to adjacent streams and rivers and under what conditions. The proposed work includes a combination of modeling and field measurements. We will begin by modeling how much (magnitude), how often (frequency), and for how long (duration) rivers are exchanging water with wetland/floodplains in the Connecticut River watershed. We will use this information to pick several wetlands and floodplains with variability in magnitude, frequency, and duration for field measurements. During the field campaign, we will use a combination of continuous instruments and water samples to determine how much CO2 is moving from the wetlands and floodplains to rivers. We will then build a computer model that captures these field measurements to estimate these exchanges for the entire watershed. Using this model, we will then test the overall importance of magnitude, frequency, and duration of flooding to CO2 concentrations and emissions from streams and rivers of different sizes and during different seasons. This work will improve understanding of the contribution of aquatic habitats and their connectivity to greenhouse gas emissions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2026 · 2024-09

Despite the high prevalence of pregnancy loss during the early post-implantation period, a detailed understanding of cell and molecular workings at this stage of development remains elusive. During this period, specification of the principal lineages of the future body occurs at the gastrulation stage, an evolutionary conserved landmark event in life. While early embryonic cells are known to be sensitive to the changes in the metabolite availability in their immediate surroundings, extremely little is known about the role of the intrauterine metabolic environment during gastrulation and how it shapes embryo viability. The maternal metabolic environment can be disrupted via somatic mutations in metabolic enzymes, such as the gain-of-function mutation IDH2R140Q. This mutation leads to the conversion of the tricarboxylic acid cycle metabolite alpha-ketoglutarate (αKG) into the epigenetically active metabolite 2-hydroxyglutarate (2-HG), which subsequently accumulates in the bloodstream of affected patients and has downstream metabolic effects. In my preliminary work, I have modeling maternal metabolic dysfunction by inducing this mutation in results in significant developmental delays at the time of gastrulation and failure to form distinct primary germ layers. My work also revealed increased histone methylation, as well as differential expression of genes involved in key developmental processes, such as cellular migration, as a response to 2-HG exposure in 2D cell culture. Collectively, these findings suggest that maternal metabolic dysfunction driven by mutant IDH is prohibitive to proper gastrulation. In light of these findings, I hypothesize that maternal 2-HG accumulation disrupts primary germ layer formation via both bioenergetic and epigenetic mechanisms. My first aim is to characterize the impact of IDH2R140Q-driven maternal metabolic dysfunction on primary germ layer formation. I will characterize the morphological effects of maternal 2-HG accumulation using high-resolution 3D confocal microscopy to investigate the spatiotemporal dynamics of germ layer cell specification and expansion. I further will characterize the changes in mitochondrial activity and cell death caused by maternal 2-HG accumulation using fluorescence-based assays. My second aim is to evaluate changes in the embryonic epigenetic landscape caused by maternal 2-HG accumulation. I will identify variable histone modifications in exposed embryos and the associated genomic loci using histone modification profiling followed by Cleavage Under Targets and Tagmentation (CUT&Tag) in the embryonic portion of gastrulas. Together, this project will pave the way toward a mechanistic and functional understanding of how maternal metabolic dysfunction modulates embryonic development as well as adverse pregnancy outcomes. Thus, in addition to providing me with valuable training that will further my career as a developmental biologist and reproductive medicine specialist, the proposed research has significant potential to provide a rich source of new molecular and cellular targets for therapeutic intervention in clinical settings where embryonic development is compromised.

NIH Research Projects · FY 2024 · 2024-09

PROJECT SUMMARY As modern human expanded worldwide, they settled in different environments that subjected them to selective pressure, driving genetic adaptations to varied local conditions. The origin of modern adaptive variants is intricately intertwined with population history and local adaptations that shaped the phenotypic diversity among human population today including health related traits. The investigation of the functions of adaptive genetic variants provides important insights into the mechanisms of human evolution and facilitate the identification of complex disease genes. Nonetheless, current strategies for establishing connections between genetic variants and functions face several limitations and new approach are needed. This proposal describes my plan to use ancient human DNA to explore the origin of adaptive variants and the strength of selection that has influenced their occurrences over time. Up to now, the use of ancient DNA (aDNA) to investigate adaptation events was challenged by the scarcity of human fossils. To overcome this barrier, I leverage on a new approach, I and others pioneered, to retrieve human aDNA from archaeological sediment independent of skeletal remains. This breakthrough technology allows us to generate genomic time-series data at a previously unachievable scale in any locations where human have once lived. I will apply this innovative approach to investigate the evolution of recent human phenotypes in East Asia. I propose to extract aDNA from a collection of sediment samples from the Tsagaan Agui cave, an archaeological site in Mongolia with evidence of ancient human occupation spanning the last 500,000 years. Mongolia, with its rich archaeological records is an ideal location not only to retrieve aDNA from modern humans but also from now-extinct archaic humans that once thrived in the region and were adapted to local environmental conditions hundreds of thousands of years before the arrival of modern human. I will use target sequencing to infer the ancestry of past individuals that occupied the site and genotype them across thousands of adaptive loci. By combining these data to the archaeological and past ecological records of the region, I will contextualize the emergence of specific adaptive variants in both archaic and modern humans. This project will significantly enhance our understanding of past environmental conditions that contributed to the acquisition of specific adaptive variants in modern human populations and holds the promise of addressing fundamental queries about our evolutionary history by defining the genetic basis of human adaptation to local environments, diets, and modern diseases.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY This is a new application by an early-stage investigator with a long-term career objective of transforming cardiovascular care using artificial intelligence and data science. The proposal focuses on aortic stenosis (AS), a progressive narrowing of the aortic valve, which manifests in older adults and causes significant disability and premature mortality despite minimally invasive treatment strategies. AS is either diagnosed following symptom- driven diagnostic testing or incidentally discovered, which has simultaneously led to a vast underdiagnosis of advanced stages of AS while identifying many with early-stage aortic valve disease without clarity on appropriate follow-up. There is a critical need for novel screening and prognostication strategies for AS. We show that artificial intelligence (AI) models applied to 1-lead electrocardiograms (AI-ECGs) can be a sensitive and convenient screen for advanced (moderate/severe) AS. AI-ECG can be paired with a second, more specific, AI-enhanced handheld cardiac point-of-care ultrasound (POCUS). This AI-POCUS automates the diagnosis of advanced AS without specialized imaging or expert evaluation. In Aim 1, we propose a multicenter pragmatic RCT evaluating this 2-stage, AI-driven screening strategy for advanced AS. This innovative, technology-driven screening strategy will define a new paradigm for the efficient identification of advanced AS. In addition, we evaluate a novel strategy to bridge the critical gap in precision follow-up, especially for early-stage aortic valve disease. Early aortic valve disease – aortic sclerosis or mild AS – affects nearly a fourth of older adults over 65 years. However, there are no guideline recommendations on follow-up for aortic sclerosis, and recommendations for mild AS do not account for the substantial heterogeneity in disease progression. In our preliminary investigations from a multicenter observational cohort study, we show that a deep learning tool for echocardiographic videos – deep learning- based aortic stenosis severity index (DASSi) – can define those at substantially elevated risk of progression to advanced AS and adverse clinical outcomes. In Aim 2, we will conduct a multicenter, prospective evaluation of an individualized AS progression score among older adults with aortic sclerosis or mild AS through a protocolized Doppler echocardiogram to distinguish those with high and low rates of progression. The investigations in Aim 2 will establish the reliability of a digital biomarker for AS progression that can enable precision care and follow- up. The work is supported by the team’s broad expertise in (a) clinical medicine, including cardiology, geriatrics, and imaging; (b) technology, spanning informatics, data science, and AI; and (c) clinical trials, with experience in designing and executing studies. The evidence generated from a multicenter evaluation of low-cost AI-driven interventions can be immediately adopted and scaled to have a major public health impact. Moreover, an objective approach to the diagnosis and follow-up of AS will reduce healthcare disparities for vulnerable patients. Future work will build on these results and engage directly with communities using low-cost portable devices to improve disease detection and outcomes among those without adequate healthcare access.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY The field of human neuroimaging has long been suffering from a problem of low power, in which low signal-to- noise ratio, multiple comparisons, and small sample sizes result in insufficient statistical power for many studies. For studies attempting to reveal brain-behavior relationships via functional and structural connectomes, which are a matrix representation for statistical and physical relationships between brain regions, the story is the same. From a scientific perspective, this issue reduces the number of findings presented in the literature while also lowering the replicability of any findings. Since connectomics research strives to ultimately be clinically relevant by, for example, predicting risk for mental health conditions or informing personalized treatment approaches for those with existing illness, this problem of low power greatly hinders progress. Fortunately, recent work has introduced a framework shift in statistics whereby an emphasis on brain networks, rather than individual connections or edges, improves overall statistical power in functional connectivity analyses. Further work has shown that using information of the relationships between edges in the connectome to construct the networks results in even greater power increases, but it is not yet known whether deriving networks within-dataset is more effective than using independent networks. Therefore, this proposal will investigate whether using within-dataset networks in network-level statistical procedures results in further power increases. It will also test these network- level approaches that were developed in functional datasets on the structural connectome. In Aim 1, I will use data from 3 large functional connectivity datasets spanning several phenotypes to examine if creating the networks in the same dataset that undergoes statistical testing will offer power improvements over deriving networks from an independent dataset. In Aim 2, I will evaluate the utility of network-level statistical procedures in the structural connectome, and in Aim 3, I will extend the methodology from Aim 1 to the structural connectome to comprehensively determine whether the results seen in the functional connectome extend to the structural connectome. This work will improve understanding of which variables we can manipulate to achieve higher statistical power in human connectivity studies and lead the field towards eventual clinical relevance by improving the neuroimaging tools available to mental health researchers.

NSF Awards · FY 2024 · 2024-09

Modern Earth is covered by lush tropical forests, extensive grasslands, and soaring redwoods—in striking contrast to landscapes through much of Earth’s early history that consisted largely of bare rock and microbial mats. Plants have dramatically altered Earth’s landscape and climate (like the shapes of rivers and patterns of rainfall). However, there is currently little consensus on how the development of plants, starting with the first ground-hugging mosses and liverworts around 470 million years ago, followed by the eventual rise of trees around 380 million years ago, affected nutrient and oxygen levels both on land and in the oceans. This research combines field, laboratory, and modeling approaches to examine the effects of early land plants on the Earth system. This study focuses on the Canadian Arctic Archipelago which contains some of the best-preserved sedimentary rocks chronicling this key time period of early plant evolution. The team of researchers are studying fossil plants, pollen, and spores and geochemical elements to understand how weathering changed on land, how plant material was delivered to the ocean, how the availability of critical nutrients like phosphorus changed on the land and in the oceans, and how oxygen and sulfur levels changed in the ocean. The broader impacts activities stemming from the research include educational and mentorship opportunities for students in middle-school through graduate school. Graduate students will be co-mentored by the Principal Investigators, and undergraduates will also be recruited to analyze collected samples. The Yale Peabody Museum and the Yale Pathways to Science program will provide platforms for community-oriented outreach efforts, including educational events fostering scientific literacy and engagement in local middle-school students. The team will also take advantage of the unique opportunity provided by recent Peabody renovations to develop a new public-facing exhibit on “Ecosystem Engineering” focused on land plants and their impacts on Earth’s landscapes and ecosystems. The University of California, Riverside’s Camp Highlander program is fostering local high-school student engagement with Earth sciences. Finally, field-conducted telepresence outreach through the new “Annals of the Arctic” program, integrated with existing summer programs at Stanford, Yale, and UCR, will provide public-facing exposure to day-by-day realities of geologic fieldwork in remote terrains. This will increase the accessibility of geologic research and provide a venue for direct illustration of geologic concepts, human experiences of the dynamic nature of polar ecosystems, and their vulnerability to ongoing environmental change. Reconstructing the biotic, biogeochemical and climatic impacts of the evolution of land plants has been hampered by the commonly fragmentary and disassociated records of geochemical and paleontological change across the lower-middle Paleozoic transition, and by the limited integration of empirical observations with the mechanistic framework that can be provided by biogeochemical and Earth-system models. To address these fundamental questions, we are generating new, high-resolution field-based geochemical data (biomarker, programmed pyrolysis, carbon isotope, lithium isotope, osmium isotope, phosphorus speciation and phosphate-oxygen isotope, iron speciation, and trace metal abundances) and sedimentological and paleontological (plant body fossils, palynomorphs, graptolite and conodont biostratigraphy) records from key sections in the Canadian Arctic to reconstruct first-order ecological and environmental changes—in both continental and marine settings—concurrent with the radiation of early land plants. The Silurian–Devonian transition is an under-characterized but key interval for both land plant evolution and marine redox state, and these data will be integrated with long-term records to distinguish perturbations from more permanent state shifts. These new empirical records will be coupled to biogeochemical modeling over a range of scales—from local critical zone and seafloor diagenetic processes to continental climate and ocean and atmospheric carbon-oxygen cycle modeling—to develop a more robust process-based understanding of plant-biogeochemical feedbacks and reconstruct the long-term consequences of early land plant evolution for the broader Earth system. 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 2024 · 2024-09

Using millimeter wave observations, astronomers can probe the structure of star forming regions. Understanding the star formation process will give us better insight on how the Earth and Sun formed and will help us better understand the changes of galaxies over time. Star formation takes place in the densest parts of molecular clouds, the regions referred to as stellar cores. The investigator seeks to understand how cores form and funnel material to the stars forming in them. The investigators will use the Atacama Large Millimeter Array (ALMA) to observe dense gas surrounding forming stars at different ages. Students of diverse backgrounds, from middle-school to graduate school, will take part in this project, and will gain experience in scientific research including analyzing and interpreting radio interferometer astronomical data. This research experience will attract underrepresented minorities to the physical sciences. Another goal is to educate the public on radio astronomy, interferometry and star formation using the ALMA observations obtained during this project for exhibits and planetarium shows. Millimeter interferometer observations of star-forming regions will be used to probe the gas in envelopes within hundreds to thousands of astronomical units of the protostars in order to determine how material goes from the cloud in towards the inner circumstellar envelope and the protostellar system, as a consequence of gravitational infall, and how it is expelled out of the envelope due to the impact of protostellar outflows. The project aims to establish how infall and outflow rates change of over the lifetime of envelopes and understand the relative effect of infall and outflow at different stages of the protostars' mass-assembling and cores' mass-loss processes. The project will also establish the importance of outflows in the removal of the dense core gas and how outflows impact the infalling material as protostars evolve. The results will be used to develop an empirical model of the evolution of the mass-assembling process in low-mass protostars and to constrain numerical simulations of feedback in star-forming cores. 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 2024 · 2024-09

NONTECHNICAL SUMMARY This award supports research and education activities with a goal to develop a fundamental understanding of quantum phases of matter with many interacting particles, using the concepts of symmetry and topology. Symmetry is a fundamental aspect of a physical system and has been recognized as a powerful principle in modern theoretical physics. A familiar example is the fact that total electric charge is conserved in everyday physical processes, which can help distinguish between electric conductors, insulators and superconductors. In the realm of quantum mechanics, a system of interacting particles, such as electrons, can form unusual quantum states, in which the constituent particles move collectively in highly intricate patterns, making them resilient to small changes in external conditions. Deciphering these subtle patterns requires ideas from topology, a branch of mathematics that concerns properties of geometric shapes that are unchanged by smooth deformations. A deeper understanding of these quantum states may enable new schemes in harnessing quantum materials and designing quantum devices. This project will leverage the power of the symmetry principle to advance the knowledge of exotic quantum states. The first thrust will investigate the properties of "symmetry defects", which are specific disturbances introduced to the system to reveal its underlying symmetry. Studying how the system reacts to such changes provides new methods to observe and characterize the unique quantum properties of the states involved. In the second thrust, the focus will be on the relationship between physical properties of a quantum crystal and microscopic interactions at the atomic level. This will furnish new perspectives on how crystalline quantum materials can be modeled theoretically. In the last thrust, the project will explore how the quantum behaviors of many interacting particles are affected by their interactions with the environment. Progress on this front will be crucial in exploiting these quantum states as memory for quantum information processing. This award also supports the PI's educational and outreach activities through mentoring and training of graduate students and postdocs in theoretical condensed matter research; writing pedagogical review articles and organizing conferences and workshops; outreach to K-12 students and the general public. TECHNICAL SUMMARY This award supports research and education activities with a goal to advance knowledge about quantum phases of matter with global symmetry and many-body topology as guiding principles. Characterizing the emergence of quantum phases from complex interactions between microscopic degrees of freedom represents a key challenge in quantum science. Advances in both condensed matter physics and quantum information science have significantly broaden the scope of quantum phases and provided new settings where universal behaviors of quantum many-body systems can arise. The project has three main thrusts: 1) Systematically develop a theory for a new type of non-local observables, called the disorder operators. These operators probe the fluctuations of symmetry charges in a given spatial region of the system, whose scaling behavior contains universal information about the underlying quantum state. The team will study how subleading corrections to the disorder operator depend on microscopic details, and develop field-theoretic techniques to compute them at quantum critical points. 2) Examine new aspects of UV/IR mixing: The interplay between microscopic conditions in lattice systems and macroscopic observables will be examined using the newly developed perspective of topological defects. The project will study how anomaly of a low-energy theory manifests in lattice quantum numbers and, conversely, how anomalous fractonic symmetries constrain low-energy dynamics. 3) Symmetry breaking and topological order in open quantum systems: the research team will investigate a new kind of symmetry breaking in mixed-state quantum phases and its implications on equilibration dynamics, and explore many-body topological order in quantum states under decoherence and beyond. This award also supports the PI's educational and outreach activities through mentoring and training of graduate students and postdocs in theoretical condensed matter research; writing pedagogical review articles and organizing conferences and workshops; outreach to K-12 students and the general public. 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 2024 · 2024-09

PROJECT SUMMARY/ABSTRACT Timely and accurate clinical decision-making is critical for the quality of healthcare delivery, impacting everyone from individual patients to entire public health systems. Clinicians often raise questions in their practice for decision-making (averaging two questions for every three patients seen), but rarely have time or resources to get evidence-based answers, leading to sub-optimal patient care decisions and even diagnostic error. This is particularly true for emergency departments (EDs) with chaotic, time-pressured, and high-stakes decision environments. Artificial intelligence (AI) driven question-answering (QA) systems can fill this gap, by providing real-time answers and predictive analytics, aiding clinicians in timely, accurate decision-making. Addressing this critical need, the rise of Large Language Models (LLMs), offers a transformative approach to understand complex questions and generate human-like responses. Despite their promise, two critical issues hinder the adoption of LLMs in clinical practice. The foremost challenge is their unreliability. LLMs can generate incorrect medical information, which has devastating outcomes such as misdiagnosis. The second hurdle is the lack of transparency. Many of these systems produce answers without providing reasoning and justification, making their responses less useful and undermining the trust of clinicians. The overall objective of this proposal is to develop and validate a clinically reliable and transparent LLM-based QA system and translate it into a clinical chatbot for clinical decision support, providing clinicians with accurate evidence-based information in high-stakes scenarios like EDs. During the K99 phase, I will develop novel clinically accurate LLMs (CliniGPT) with multi-modality clinical data guided by the clinical-specific pre-training and fine-tuning framework (Aim 1). During the R00 phase, I will develop and validate the retrieval-augmented medical QA (CliniQARet) framework, to guide CliniGPT in generating reliable answers to clinical questions in the ED setting (Aim 2). Using the best model from Aim 1 and Aim 2, I will build the clinical chatbot following user-centered principles, delivering evidence-based, timely support for common ED scenarios including chest pain, headache, fever, and abdominal pain, to enhance decision-making. I will develop and validate the software in a simulated EHR environment using real patient data and recruiting ED clinicians (Aim 3). The expected outcomes are a real-time, user-centered ED clinical chatbot; open-source clinically accurate LLMs; an open-source reliable and trustworthy clinical QA framework; an open-source framework for pretraining, fine-tuning, and evaluating clinical LLMs focusing on reliability; an open-source framework of constructing and integrating multi-modal clinical datasets to enrich and ground the system’s clinical knowledge. During the K99 phase, the PI will be mentored by experts in clinical NLP and LLM, emergency medicine, and clinical informatics, and requires additional training in clinical, evidence-based and emergency medicine. This application will provide the necessary training to supplement the PI’s expertise in clinical NLP and clinical medicine and help her transition into an independent career in biomedical data science.

NSF Awards · FY 2024 · 2024-09

Active dynamics encompasses a wide range of collective behaviors exhibited by flocks of birds, schools of fish, layers of cells, and networks of filamentous proteins. In all these examples, out-of-equilibrium organization emerges spontaneously from interactions among active agents that consume energy from an internal reservoir or derive it from their surroundings. This project supports the development of scalable and automated software to simulate active dynamics and to infer the laws that govern it from the analysis of state-of-the-art experimental data, such as high-resolution microscopy videos. The goals of this project are to empower researchers to reliably extract hidden rules from noisy experimental observations of active dynamics, lower the barrier for analyzing large microscopy videos, reduce the time-consuming reimplementation of simulation and estimation, and promote cross-disciplinary collaborations. The dynamics of active matter, such as collections of fibroblasts or epithelial cells, is intrinsically stochastic and out of thermal equilibrium, and affected by a variety of complex processes, such as cell division. The large intrinsic fluctuations present in active matter systems hinder the efficient extraction of signals from noisy experimental data and thus it urgently demands the development of data-science-enabled tools to accelerate the analysis and improve the reproducibility of the findings. The development of the Dynamics Lab ecosystem addresses this urgent need by establishing two classes of interconnected software packages for real-space and scattering-based analysis of microscopy data. Together, these tools enable visualization and integration of physics-based simulations and statistical machine learning in both real space and Fourier space. Furthermore, to address the practical concerns of data sharing, such as size limit, this project supports the development of an efficient paradigm for data acquisition of active dynamics, where large raw data are stored offline, and small online data sets that sufficiently capture the raw data set are easily transferred and used for most research purposes. A major goal of Dynamics Lab infrastructure is to achieve sustained impacts on basic and applied sciences, ranging from biophysics to biomimetic materials. This award by the NSF Office of Advanced Cyberinfrastructure is jointly supported by the Division of Materials Research and the Division of Mathematical Sciences. 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 2024 · 2024-09

Project Summary The central hypothesis of my F99-phase proposal is that the lack of T-cell infiltration into human tumors may be caused by the presence of tumor-derived T-cell excluders (TCEs) in the TME, which are overexpressed in tumor tissues and impair the function of T cells including their ability to migrate into tumor tissues. This provides a potential explanation why both spontaneous and therapeutic T cell-mediated anti-tumor immunity fail frequently in patients with immunologically “cold” tumors. By employing a secretome-wide in vitro T-cell transwell migration high-throughput screening (HTS) platform, we tested over a thousand soluble human proteins on the migration of activated T cells towards chemokine signal. SLIT2 was identified as a candidate target and validated to directly inhibit T-cell chemotaxis towards CXCL11, a common chemokine found in human tumor tissues. The N-terminal fragment of SLIT2 was first confirmed to mediate T-cell chemotaxis inhibition. Detailed functional domain mapping demonstrated that the first two leucine-rich repeat (LRR) domains, where the canonical receptor ROBO1 binds, are dispensable to SLIT2’s function in regulating T-cell chemotaxis. Meanwhile, ROBO1 expression could not be detected on T-cell surface with FACS staining, and soluble ROBO1 extracellular fragment fusion protein failed to neutralize SLIT2’s inhibitory effect on T cells, collectively suggesting a novel T- cell specific SLIT2 signaling axis independent of ROBO1. In a syngeneic mouse pancreatic cancer model, dramatically elevated T-cell infiltrated was observed in SLIT2 knock-out Pan02 tumors. SLIT2-KO tumors were rejected by the immune-competent C57BL/6 mice, while the SLIT2 wild-type tumors grew. Such difference diminished when the SLIT2-WT/KO tumors were inoculated into immune-compromised NSG mice. Anti-SLIT2 monoclonal antibody 11C8 neutralized the inhibitory effect of SLIT2 on T-cell chemotaxis in vitro, and its single- agent treatment led to Pan02 tumor regression in vivo. The functional SLIT2 receptor expressed on T cells, the broader impact of tumoral SLIT2 to the immune contexture within the TME, and the potential synergistic effect of anti-SLIT2/anti-PD-1 combo are subject to further investigation during the F99 phase training. In the K00 phase of this proposal, I would like to expand my study to understand dysfunction of T cells during tumor progression. These studies might include the discovery of additional TME-specific extracellular factors that drive dysfunctional anti-tumor immunity in T-cell inflamed tumors. I propose to establish in vitro HTS platforms to identify candidate targets that (1) promote central memory T-cell phenotype that favors egression into secondary lymphoid organs, (2) reduce T-cell proliferative capacity and induce apoptotic signatures, (3) suppress T-cell effector functions and drive exhaustion phenotype. Multi-omics studies and inducible expression animal models will be utilized in parallel for target identification and validation purposes. Collectively, this proposed research will extend our understanding of tumor immune evasion mechanisms, and lead to the discovery of novel immunotherapy strategies for cancer patients.

NIH Research Projects · FY 2025 · 2024-09

PROJECT SUMMARY/ABSTRACT Alcohol directly stimulates the hypothalamic-pituitary-adrenal (HPA) axis, a primary stress system, initiating the release of glucocorticoid hormones. Preclinical work indicates that chronic alcohol exposure increases both brain and peripheral corticosterone in rodents, with some studies suggesting greater increases in the brain. These increases in rodent brain glucocorticoid levels are prolonged and persist long after plasma corticosterone returns to baseline. Thus, brain glucocorticoid dysfunction may underlie long-lasting, persistent alcohol use. Studies of HPA axis dysregulation in humans with alcohol use disorder (AUD) have only examined peripheral cortisol in relation to alcohol and with mixed results. Chronic alcohol use is associated with both elevated and blunted levels of peripheral cortisol in response to alcohol and stress in individuals with AUD. So, what we know about HPA dysregulation in AUD is based solely on peripheral cortisol, and results are inconsistent and incomplete because we have not been able to study a marker of brain cortisol - forming a significant gap in the existing literature. Translating preclinical findings in the brain to human AUD and examining the relationship between brain-peripheral cortisol is a critical next step in understanding HPA dysfunction in this population. Levels of glucocorticoids in the brain are dependent on the enzyme 11β- Hydroxysteroid dehydrogenase type 1 (11β-HSD1), which catalyzes the conversion of cortisone to cortisol (or corticosterone in rodents). 11β-HSD1 is expressed in brain regions critical to alcohol addiction, including the amygdala and prefrontal cortex (PFC). Using a novel radiotracer called [18F]FMOZAT together with Positron Emission Tomography (PET) brain imaging, we can now measure levels of 11β-HSD1 in the living human brain. Using these methods, we have exciting preliminary data (obtained through K01AA025670) suggesting that individuals with AUD (n=9) have higher 11β-HSD1 availability compared to healthy controls (n=12) in prefrontal-limbic regions. Preliminary data also demonstrate that 11β-HSD1 availability in ventromedial PFC is associated with drinks per week, quantity of drinks per drinking episode, and AUD severity. However, we do not know how 11β-HSD1 availability in brain corresponds to peripheral cortisol in humans, and whether brain vs. periphery predicts drinking behavior. This proposal will derive critical foundational information on brainperiphery HPA axis dysregulation by: fully characterizing both availability of 11β-HSD1, a cortisol regenerating enzyme, in brain and peripheral cortisol in those with AUD vs. healthy controls, and determining whether brain vs. periphery predicts drinking behavior (Primary Aim 1). We will also evaluate whether 11β-HSD1 availability vs. peripheral cortisol predict stress-related drinking using EMA and a biosensor system in individuals with AUD compared to healthy controls (Exploratory Aim 2). This will build the future foundation for additional mechanistic studies probing the HPA axis and provide needed data to support translational upstream HPA axis targets as AUD treatments in humans.