Stanford University

NIH Research Projects · FY 2025 · 2025-08

Project Summary Cumulative evidence indicates neuromyelitis optica (NMO) patients suffer from disabling cognitive decline (CD) despite effective immune therapies. We must therefore identify modifiable risk and protective factors to prevent neurologic disability from CD. Indeed, there is high interest among people living with CD if immune therapy preserves cognitive function. We have shown in our longitudinal, multi-center study using Montreal Cognitive Assessment (MoCA) that up to a third of NMO patients suffer CD, independent of clinical relapse. This R21 will evaluate the pathogenic NMO-IgG as a mediator of CD in NMO. Distinct patterns of CD among NMO patients exist, but the underlying mechanism is unknown. Others have shown brain atrophy occurs in NMO even in the absence of cerebral syndromes, while we have linked the NMO-IgG index titers with brain atrophy, which informed our hypothesis that NMO-IgG is a mediator of CD by targeting the blood- brain barrier (BBB) astrocytes with resultant gray and white matter dysfunction. Using NMO as an experimental model of immune-mediated CD, this study will leverage two distinct pre-existing cohorts with rich clinical and imaging data accompanied by biospecimens. The first is CIRCLES (Collaborative International Research in Clinical and Longitudinal Experience Study) cohort comprised of >1000 NMO patients from 2013-2020. This current proposal will link in-depth cognitive analyses with high-resolution, state-of the-art imaging data and cutting-edge proteomic blood biomarker studies, while mining over 8 years of follow-up data. The second is a prospective, ongoing biorepository, Project BIG (Stanford Brain, Immune and Gut Initiative), accessing data from >100 NMO patients, which will provide additional information on treatment effects on cognition over time. Specific aims are designed to test the hypothesis that NMO-IgG expression is directly linked to BBB deficiency and white matter pathology, leading to CD (aim 1), and that astrocyte damage from the NMO-IgG attack leads to gray matter dysfunction and CD (aim 2). Detailed data on clinical and socio-demographic modifiers of cognitive function will be accounted for in all analyses. The immediate impact of this study is to stop subclinical, relapse-independent progression and enhance productivity and quality of life, eagerly sought by patients living with NMO. In broader terms, this proposal has a high potential for advancing our understanding of the pathophysiology of immune-mediated CD and development of therapies targeting neurotoxic immune responses.

NIH Research Projects · FY 2025 · 2025-08

Project Summary Astronauts often experience health challenges after spaceflight due to the body's adaptations to a microgravity environment. This UG3/UH3 project aims to explore how spaceflight impacts heart health, specifically focusing on the changes that occur in human heart under microgravity. Using stem cells, we will create 3D heart-like tissues, known as vascularized cardiac organoids (vCOs), which contain both heart cells and blood vessel cells. These organoids will be sent to the International Space Station (ISS) to study the effects of spaceflight on their function over time, including changes in their ability to beat, metabolize energy, and respond to inflammation. Our primary goals are to identify the biological pathways disrupted by spaceflight and to evaluate potential drugs that could protect the heart from these effects. We will use single-cell RNA sequencing, which allows us to observe how individual cells respond to spaceflight, combined with computational drug screening to identify promising therapeutic candidates. By comparing the responses of organoids in space with those grown on Earth, we aim to gain insights into how microgravity and space radiation affect heart function and identify new strategies for maintaining heart health during extended space missions. The findings from this study will improve astronaut safety during long-duration missions, such as those to Mars, by identifying effective countermeasures against space-induced cardiovascular changes. Additionally, the research could have broader implications for understanding heart health on Earth, potentially benefiting individuals with heart disease or inflammatory conditions. Our work seeks to bridge the gap between space-based and Earth-based heart research, offering new perspectives that could lead to innovative therapies for sustaining cardiovascular health in challenging environments.

NIH Research Projects · FY 2025 · 2025-08

X-linked adrenoleukodystrophy (ALD) is a devastating genetic disorder, caused by mutations in a peroxisomal gene (ABCD1), in which two-thirds of affected males develop cerebral ALD (cALD), a progressive, inflammatory brain condition that is often fatal. Currently, the only available treatments are hematopoietic stem cell transplantation (HSCT) and ex vivo gene correction (FDA-approved in 2022). However, these therapies are limited to a subset of cALD patients based on factors such as lesion stage, age, donor availability, and access to advanced healthcare facilities. Consequently, most boys and men worldwide who develop cALD die from it. We propose using a novel cALD mouse model to test a first-in-class, monoclonal antibody therapy targeting fibrin, a potent pro-inflammatory protein that is highly expressed in cALD brain lesions. Fibrin, a hallmark of blood-brain barrier (BBB) disruption, plays a crucial role in neuroinflammation and demyelination by activating macrophages, microglia, and astrocytes, and impairing oligodendrocyte maturation. Our approach leverages our recently developed mouse model which combines cuprizone (CPZ) diet and MOG injection (EAE model) in Abcd1-knockout mice to induce a cALD phenotype. Our model replicates key features of human cALD, including demyelination, axonal damage, BBB disruption, oxidative stress, and fibrin deposition. We will evaluate the efficacy of 5B8, an antibody targeting fibrin's pro-inflammatory epitope (amino acids 377-395), in alleviating neurological disability and reducing pathological manifestations in cALD mice. We will compare the effect of therapy in early (presymptomatic) and late (symptomatic) stages of the disease and against sham IgG control. The project comprises three aims: In Aim 1, we will investigate whether anti-fibrin immunotherapy ameliorates behavioral symptoms by assessing motor function using the open field test and cognitive function using the Barnes maze. In Aim 2, we will examine whether anti-fibrin immunotherapy improves BBB disruption in the brain using T1- weighted MRI and increases cerebral blood flow using Arterial Spin Labeling MRI in the cALD mouse model. We will also analyze plasma biomarkers of BBB disruption and neuroinflammation, including neurofilament light chain, interleukin-18, vascular cell adhesion molecule 1, and other markers. In Aim 3, we will confirm target engagement and evaluate whether anti-fibrin therapy reduces histological brain lesions by performing immunostaining to measure BBB integrity, immune cell infiltration, and axonal damage. The successful completion of these aims will generate necessary preclinical data and identify relevant biomarkers to facilitate the design of a Phase 2/3 human clinical trial for anti-fibrin immunotherapy in cALD. Success could lead to a novel treatment option for this devastating disorder, addressing a serious unmet medical need and offering hope for patients currently ineligible for existing therapies.

NIH Research Projects · FY 2026 · 2025-08

PROJECT SUMMARY First designed to treat diabetes, sodium-glucose cotransporter-2 inhibitors (SGLT2i) were used to prevent glucose reabsorption in the kidney. Recent clinical trials of SGLT2i further demonstrated an unexpected and substantial reduction in heart failure hospitalizations in patients with and without diabetes. Since SGLT2 is lowly expressed in the heart, its off-target mechanisms present a fascinating opportunity to elucidate cardiac protective targets beyond glycemic control. This K99/R00 application describes a five-year research training plan that leverages (i) human induced pluripotent stem cell-derived cardiovascular cells, (ii) single-cell RNA transcriptomics (scRNA-seq), (iii) metabolomics, (iv) large-scale drug-protein interaction determination, and (v) cell and animal validation to elucidate cardiac-protective mechanisms in health and disease. Given the well- established, off-target protective mechanisms of SGLT2i in the heart, the applicant, Dr. Arianne Caudal, will test the hypothesis that SGLT2i promotes mitochondrial biogenesis and metabolic remodeling, maintaining energy homeostasis in heart failure. In Aim 1 (K99), Dr. Caudal will use a “cell village” multi-omic population screening platform to determine the transcriptomic and metabolomic response conferred by SGLT2i in cardiomyocytes (iPSC-CMs), fibroblasts (iPSC-CFs), and endothelial cells (iPSC-ECs). In Aim 2 (K99), Dr. Caudal will determine the direct protein binding partners of SGLT2i using a cutting-edge proteomics approach in three-dimensional iPSC-derived engineered heart tissues (EHTs). In Aim 3 (R00), Dr. Caudal will validate mitochondrial pathways using pharmacological induction of cardiac dysfunction in iPSC-CMs and a mouse model of pressure overload- induced hypertrophy heart failure. Furthermore, these methodological pipelines provide a springboard of applicability to a range of small molecules, metabolites, and peptides, creating a systems biology niche for Dr. Caudal’s independent work. The proposed studies build upon PI Dr. Arianne Caudal’s well-suited prior training in iPSC modeling, proteomics, and mitochondrial metabolism while providing new training opportunities in (i) precision health, (ii) single-cell multi-omics, and (iii) animal modeling. Mentor Dr. Joseph Wu is a pioneer in iPSCs and cardiovascular biology, and co-mentor Dr. Michael Snyder is a leading expert in single-cell multi- omics and precision medicine. Collaborators and advisory committee members Drs. Zoltan Arany (cardiac metabolism, heart failure), Devin Schweppe (protein interactions), Allis Chien (mass spectrometry), and Sarah Heilshorn (tissue bioengineering) provide additional expertise and guidance. In summary, the well-tailored research training plan, exceptional mentoring team, and outstanding environment at Stanford University are anticipated to help propel Dr. Caudal toward her long-term goal of establishing an independent research program at the intersection of cardiovascular metabolism and systems biology.

NSF Awards · FY 2025 · 2025-08

This project develops combinatorial mathematics for new biological concepts that have emerged from genomic studies of phylogenetic trees and networks. Phylogenetics—the interpretation and construction of trees and networks that describe relationships of shared descent from common ancestors—is central throughout the entire field of biology, contributing to biological subfields in areas as varied as developmental biology, ecology, evolution, genetics, microbiology, and the biology of cancer. The research advances the fundamental understanding of mathematical structures that underlie descent relationships of cells, species, and strains for multifarious biological applications. The project develops the mathematical area of phylogenetic enumerative combinatorics at the intersection of mathematics and biology, toward performing classifications of new discrete combinatorial structures that relate to phylogenetic trees and networks. The project performs unified enumerative studies of several objects, advancing an approach for mathematical analysis of novel phylogenetic combinatorial structures. The structures that will be investigated include: the “perfect phylogenies” that appear in contexts involving genetic recombination and algorithmic improvements in phylogenetic computation; the “galled trees” and “rankable tree-child networks” that assist in extending trees to include biological processes of hybridization and horizontal gene transfer; the “ancestral configurations” that arise from consideration of genetic lineages descending through speciation events; and encodings of trees for phylogenetic analysis of pathogen strains. Its combinatorial analysis proceeds by a shared multidisciplinary framework linking biological structures with methods from the fields of enumerative combinatorics, analysis-of-algorithms, and analytic combinatorics, employing lattice theory, recurrences, generating functions, asymptotic growth analysis, and correspondences with existing combinatorial structures. The project strengthens linkages between mathematics and biology, advancing the mathematics that undergirds the field of phylogenetics, an area that itself serves as a central unifying topic throughout biology. Further, it promotes interdisciplinary mathematical and biological training at the postdoctoral and PhD levels and advances undergraduate research at the interface of biology and mathematics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-08

PROJECT SUMMARY I am a molecular immunologist fascinated by T cell immunology. My scientific interest is applying molecular and computational approaches to study the human T cell repertoire and antigenic epitopes during infection. In prior research, I acquired skills and knowledge that are highly aligned with my interests, which resulted in consecutive first-author publications in prestigious journals, including Immunity, Cell Reports, and Frontiers in Immunology. My short-term goal is to become an assistant professor and start my independent research. My long-term goal is to become a leading scientist in T cell recognition. Stanford University provides the best possible training environment through world-leading faculty, facilities and resources to fulfill my career goals. My postdoc mentor, Dr. Mark M Davis, has been a world-leading T cell immunologist since 1980s. He has mentored over 70 postdocs and most are on tenure-track positions at renowned universities. My career training will be under the close guidance of Dr. Davis as well as my Advisory Committee with five outstanding senior professors from Stanford, Yale and MIT, who are Drs. Calvin Kuo, Akiko Iwasaki, Darrell Irvine, Benjamin Pinsky, and Holden Maecker. Each of them and I have developed an individualized research training and career development plan. My training will focus on advanced methodologies and knowledge, academic activities, faculty job hunting, grant application, scientific publication, laboratory management, and long-term collaborations. This proposal will focus on the identification of protective T cell epitopes that have potential to formulate HPV therapeutic vaccine. Most vaccines prevent viral infection effectively by inducing neutralizing antibodies, but they fail to eliminate pathogens post-infection, one such example is the existing HPV vaccines. It’s striking that while most HPV-infected females clear the virus naturally, many do not and progress to cervical cancer. T cells are the complements of antibody responses to maintain host immune competence. We recently found that there is a small fraction of T cell specificity groups uniquely enriched in females who cleared HPV. It is worth noting that this pattern is similar to what our group found in the control of Mtb (Nat Med, 2023). Thus, I hypothesize that there are unique T cell specificity groups in HPV-infected females that correlate with HPV clearance, and T cells that contribute to clearance are subdominant during infection. This is a critical question as prior studies on protective T cells largely focus on dominant T cells. The identification of T cell epitopes is technically challenging, especially for those recognized by subdominant T cells. Moreover, as the T cell response is MHC restricted, it is a longstanding barrier to probing T cell epitopes with appropriate experimental models. I am proposing to tackle these problems with cutting-edge technologies, this study will comprehensively cover T cell receptor (TCR) sequencing, protective TCR discovery, T cell epitope identification and the establishment of probing models. The novel concepts and the knowledge generated in this study will pave a path to identify protective T cell epitopes for T cell-based vaccines, not only for combating HPV infection but also for viral infection in general.

NIH Research Projects · FY 2025 · 2025-08

Project Summary/Abstract The outlined proposal expands the previous research in preterm infant brains by integrating multiple advanced neuroimaging methods, including novel quantitative magnetic resonance imaging (MRI) techniques and environmental factors to predict long-term neurodevelopmental issues. Preterm infants have a high rate of long-term neurodevelopmental impairments that often require additional health care and intervention. Recent advances in neuroimaging have provided great insight into the patterns of specific alterations in preterm infants. However, the studies are limited to demonstrate how brain function and structure and their interaction relate to or explain later neurodevelopmental outcomes in preterm infants. Further, while recent studies showed that microstructural tissue properties using recently developed quantitative MRI techniques (multi-shell diffusion tensor imaging, quantitative T1, qT1) are more sensitive and reliable to define underlying neurological diseases and, thus, it is crucial to investigate the microstructural tissue properties during the neonatal period to fully understand how these measures are related to preterm birth and predict later cognitive problems in preterm infants. The overarching aims of this project are designed to use neonatal multimodal neuroimaging techniques, including qT1, for the first time, combining with environmental factors using advanced computational approaches to define early biomarkers of later preterm neurodevelopmental outcomes. We hypothesize that the links between structural-functional brain networks are significantly altered, and when combined, will provide an exclusive prediction on later neurodevelopmental outcomes. Similarly, we will test the working hypothesis preterm infants will have abnormal white and grey matter microstructures, and these patterns will be correlated with the neurodevelopmental problems. Furthermore, we will explore the environmental factors contributing to later cognitive issues that will play an essential role in predicting later neurodevelopmental problems in preterm infants. The proposed research results will provide an early diagnostic tool that could inform the treatments and implementation of preventative interventions before any cognitive problem emerges. It also has an important impact on identifying behavioral targets to improve the life course outcomes in preterm infants.

NSF Awards · FY 2025 · 2025-08

Large-scale brain recording technologies now enable the activities of thousands of neurons to be monitored simultaneously in freely moving animals. In parallel, advances in computer vision enable fast and accurate tracking of animal behavior, which offers rich opportunities for evaluation of the nervous system's output. Capitalizing on these new technologies, this project focuses on critical statistical challenges that must be overcome to tackle a central challenge of neuroscience: linking brain activity to behavioral output. Neural and behavioral data are noisy, high-dimensional time series with complex dynamics. This project will develop novel computational and statistical methods to analyze these data and offer new insight into how the brain produces natural behavior. Specifically, this project will develop state space models (SSMs) — statistical models for high-dimensional time series data — to link brain activity and behavior. The first objective is to develop SSMs for neural recordings. These methods will allow scientists to model how brain activity changes across contexts, how it is influenced by chemical signals like neuromodulators and neuropeptides, and how it can be precisely modulated through real-time, closed-loop experiments. The second objective is to develop similar methods for the study of behavioral video data, allowing scientists to quantitatively capture how behavior is shaped by modulatory inputs, how it changes throughout aging, and how behavior is structured in pursuit of latent goals. Finally, this project will leverage state-of-the-art methods in artificial intelligence called deep SSMs, tailoring them for use in brain-computer interfaces, and developing a theory of how deep SSMs work. Altogether, this project will contribute novel methods that will provide new insight into how the brain produces rich, natural behavior. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-08

Abstract: Over 8 million people in the US suffer from peripheral artery disease (PAD), which is characterized by narrowing of the arteries in the arms or legs that lead to insufficient blood flow and limb ischemia. A therapeutic strategy to treat PAD is to boost the formation of new vessels through a process known as angiogenesis. We previously demonstrated that the delivery of human induced pluripotent stem cell-derived endothelial cells (iPSC-ECs) improve blood perfusion and angiogenesis in preclinical models of PAD. However, a major bottleneck is poor post-transplantation iPSC-EC survival and the limited capability to reshape the local immune response. To overcome these limitations, we propose to develop novel synthetic biology strategies that would enable the quantitative and dynamic regulation of growth factors and cytokines produced from iPSC-ECs after transplantation. Specifically, the strategies leverage a novel control knob we developed from a human protease and its FDA-approved inhibitor, which would facilitate its eventual deployment in patients because of its reduced immunogenic risk compared to existing control tools and the availability of off-the-shelf external control. Specific Aim 1 is to engineer versions of a pro-survival growth factor (bFGF) and two immunomodulatory cytokines (IL-10 and IL-19) that can be controlled by the protease. We will engineer variants for protease-dependent activation and inactivation, achieved by fusing the target proteins to “caging” domains that are removable by the protease and inserting protease cleavage sites into internal loops, respectively. Specific Aim 2 is to test the pro-survival effect of the protease-controlled growth factor on iPSC-ECs in vivo. We will achieve compact encoding of the entire system on a single vector and validate its performance in vitro. We will then transplant the modified cells in a mouse model of PAD. Output measures include bioluminescence to track cell survival and laser Doppler spectroscopy to quantify vascular perfusion recovery. Specific Aim 3 is to examine the immunomodulatory effects of the protease-controlled cytokines in vivo. After compact encoding and in vitro testing similar to Aim 2, we will transplant the engineered cells in vivo and perform immunohistochemistry to quantify how they affect the immune response at the transplantation site. These results will not only pave the way for more effective stem cell therapies and immunomodulation of the ischemic limb environment for the treatment of PAD, but also provide the first proof of principle for a control strategy that could be generalized to other cell therapies.

NIH Research Projects · FY 2025 · 2025-08

Abstract Chronic musculoskeletal (MSK) pain affects the lives of over a quarter of youth and impacts multiple domains of functioning, including social, emotional, and behavioral functioning. Digital behavioral health interventions offer solutions to existing access to care barriers, with outcomes similar to in-vivo treatment. Despite this, only 28% of digital tools are disseminated adequately or timely. Most research on digital interventions has focused on evaluating efficacy and adoption in the context of clinical trials but have largely ignored service design, the process in which patients and clinicians engage with each other and with technologies when integrating an intervention into practice. Ignoring the users or contexts (e.g., clinics) in which interventions are implemented leads to suboptimal healthcare innovation and significant research waste. Given this, there is an imperative need to integrate service design methods as part of the development of digital behavioral interventions for youth with chronic MSK pain. With my K23 Mentored Patient-Oriented Research Career Development award, I am designing and evaluating feasibility and preliminary efficacy of iGET Living, a digital graded exposure treatment (GET) for youth with chronic MSK pain. My long-term goal for this program of research is to establish iGET Living as an evidence-based intervention that is scalable and sustainable and can be broadly implemented in clinical care. To achieve this goal, I need to design the intervention (K23), understand how to integrate it into clinical care (R03), and validate its effectiveness and implementation in a hybrid trial (R01). The overall objective of this R03 small grant program application is to apply service design methods to understand how to implement iGET Living in three clinical settings. I will partner with three clinics that are typical settings youth with chronic MSK present and are the planned sites of a subsequent R01: pain management, rheumatology, and orthopedics/sports medicine. Using the 4-Phase Double Diamond Model design process, I will apply service design methodologies and collaborate with the clinics to create two products that describe how iGET Living will be embedded in each setting: Service blueprints and implementation roadmaps. Service blueprints are diagrams of the touchpoints when person-to-person and person-to-technology transactions occur through the intervention delivery process. Implementation roadmaps specify the implementation strategies and adaptions to the intervention and settings that are needed to support implementing the service blueprints for iGET Living. Dedicated attention to service design in advance of an R01 investigation is ideal for: 1) strengthening partnerships with the clinics for longer-term collaboration; 2) accelerating the timeline to deliver the intervention to patients by completing the implementation-preparation activities; and 3) identifying ways to address potential challenges when conducting research in real-world settings so they can be later avoided. Designing for the delivery of iGET Living for pediatric MSK pain will accelerate my science from K23 to R01 effectiveness- implementation investigation and propel digital intervention research toward improved engagement and impact.

NIH Research Projects · FY 2025 · 2025-08

PROJECT SUMMARY: Surgical pain is caused by tissue injury and inflammation. To treat surgical pain, people are given opioids which is leading to secondary health problems after surgery including opioid abuse, dependence, and overdose that is driving the United States opioid epidemic. This is particularly important for older adults as treating pain with opioids also present challenges with age-related changes in drug metabolism in addition to drug-drug interactions due to polypharmacy. Thus, discovering new non-opioid targets and developing non- opioid treatments that are effective in the young and elderly to alleviate surgical pain are urgently needed. For this HEAL R03 proposal, the goal of this research is to further understand the role of dipeptidase 2 (DPEP2) in mediating inflammation leading to surgical pain. In the traditional pro-inflammatory role, DPEP2 converts leukotriene D4 to leukotriene E4 leading to inflammation. However, in a recently discovered anti- inflammatory role, DPEP2 also limits the activation of the canonical NF-B pathway. Here we will leverage new key findings regarding the DPEP2 catalytic site (based upon the crystal structure and functional assays for another DPEP family member) to develop catalytically active and inactive DPEP2 constructs to understand whether DPEP2 limits the activation of the NF-B pathway through a peptidase-dependent or independent mechanism. In order to carry out this work, we will assess how DPEP2 regulates the NF-B pathway by using stable NF-B luciferase reporter cell lines (skin fibroblast NIH-3T3 and macrophage Raw264.7). Further, we will determine whether targeting the cysteinyl leukotriene receptor 1 (CysLTR1) downstream of DPEP2 will limit pain and inflammation after injury using a surgical incision model. To carry out these studies, we developed a rodent paw surgical incision model that increases phosphorylated NF-B 3-fold in addition to a 10-fold increase in NF-B-regulated pro-inflammatory cytokines including IL-6 and IL-1. Taken together, this proposal can advance the field by further understanding the role for DPEP2 in regulating inflammation and whether a non- opioid therapeutic targeting CysLTR1 in the DPEP2 pathway is effective at reducing surgical pain in rodents. Additionally, since the studies performed for this proposal will determine whether DPEP2 limits NF-B activation and production of NF-B-mediated proinflammatory genes, these studies also have a broad importance to aging. This is because inflammation is a hallmark of aging and the cellular senescence- associated secretory phenotype described in aging cells is driven by increases in IL-6 and IL-1; where IL-6 is the major driver of this phenotype. Therefore, understanding further the role of DPEP2 may potentially have broader implications besides treating surgical pain and useful in limiting inflammatory pain for the elderly.

NIH Research Projects · FY 2026 · 2025-08

ABSTRACT Inherited retinal degeneration is a leading cause of irreversible blindness in children. There is currently no effective therapy for any of these diseases except RPE65. The systemic manifestations of the syndromic forms of ciliary degeneration make them a particularly challenging form of the inherited retinal degeneration family to treat. Joubert syndrome is a rare inherited form of retinal degeneration in which patients die prematurely and which has specific defects localized to the connecting cilium of photoreceptors. ARMC9 is a recently discovered gene that has been implicated in Joubert syndrome. This application seeks to identify the mechanism underlying the interplay between the loss of ARMC9 and impaired mitochondrial function in cilia formation. We have strong preliminary data supporting a novel link between mitochondrial function and ciliogenesis. We hypothesize that augmentation of mitochondrial function in ARMC9 mutations will rescue the loss of function. We aim to (1) dissect the mechanism by which mitochondrial proteins contribute to ARMC9 localization, (2) determine whether NMNAT overexpression rescues the loss of ARMC9, (3) determine whether CRISPR-mediated gene editing of ARMC9 restores photoreceptor formation in human iPSC-organoids. These studies will further our understanding of mitochondrial and ciliary function and facilitate the discovery of new therapies for ciliary disorders, including photoreceptor degeneration.

NIH Research Projects · FY 2025 · 2025-08

This proposal requests funding for a Bruker timsTOF fleX MALDI-2 mass spectrometer to support spatial single-cell omics research at Stanford University. The instrument will be placed at the Stanford Center for Genomics and Personalized Medicine (SCGPM) Genome Sequencing Service Center (GSSC). Increasing evidence from single-cell studies reveals that spatial distribution of biomolecules is crucial for understanding tissue complexity,cell to cell communication and their interaction with microenvironments. Although our facility has a robust selection of mass spectrometers for bulk omics studies, spatial single cell omics requires specialized equipment. Currently, our facility excels in supporting spatial single cell transcriptomics and proteomics, technologies widely available to our researchers. The crucial missing piece is an instrument capable of studying spatial distribution of small molecules at single cell resolution. A comprehensive understanding of cellular heterogeneity and the microenvironment requires not just RNA and protein expression, but also the distribution of small molecules such as metabolites, lipids and glycans. Correlating small molecule distribution with gene and protein expression helps establish crucial links between genetic instructions and their functional consequences. The proposed instrument offers unmatched single-cell resolution for analyzing the spatial distribution of small molecules, complementing existing spatial transcriptomics and proteomics studies. This cutting-edge instrument will offer several unique features: (1) Un-paralleled spatial resolution to single cells for tissue imaging; (2) Fast data acquisition speed which is essential for imaging large tissue areas and high throughput screening; (3) Streamlined workflow for easy operation, data visualization and analysis. Access to this instrument will facilitate multiple levels of translational and basic science research. This includes constructing molecular maps of human tissues at single-cell resolution, conducting in situ tissue glycomics analysis for biomarker discovery and disease mechanism investigation in various types of cancer, and investigating how the alternation of small molecules contributes to lung and cardiovascular disease, cancer development, aging, and more. Progress on this broad array of projects will be catalyzed by the effective usage of the new instrument through close cooperation among the user groups. By serving a highly productive interdisciplinary group of NIH-funded investigators, the proposed instrument will enhance existing research programs while encouraging new projects and collaborations to emerge and be funded.

NIH Research Projects · FY 2025 · 2025-07

ABSTRACT Pancreatic ductal adenocarcinoma (PDAC) is a frequent and almost uniformly fatal malignancy. PDAC metastasizes to many different organs and is largely refractory to conventional chemo- and immuno-therapies. PDAC cells exist in diverse cell states, however, the extent to which these states dictate metastatic ability, organ tropism, and immune evasion remains largely unexplored. The critical molecular drivers underlying metastatic fitness phenotypes that drive success at each step of the metastatic cascade are likely highly diverse and remain largely unknown. To bridge this knowledge gap, we will integrate genetic barcoding with single cell profiling to uncover the determinants of the fundamental properties of metastasis. We are uniquely positioned to conduct this multidisciplinary systems-level proposal which employs innovative methods for multiplexed dissection of the metastatic process. We will employ molecular barcoding and high-throughput barcode sequencing to quantify the metastatic ability of an extensive panel of >30 PDAC cell lines across time, organ site, and immune environment. Our preliminary data from our pooled and barcoded cell line panel demonstrate highly heterogeneous metastatic phenotypes and cell states, and that our approach provides both cell line-level and clonal resolution of metastatic ability. In Aim 1, we will pool our barcoded cell lines for several types of transplantations that mimic modes of metastatic spread and quantify the metastatic ability of highly diverse cell states across different time durations, organ sites, and immune environments. Similar approaches will be employed to investigate cancer adaptation for immune evasion. In Aim 2, we will relate metastatic phenotypes to the molecular programs underlying diverse cell states and determine how cell state influences metastatic fitness and immune evasion. By performing scRNA-seq on the multi-component barcode in our cell lines following intrasplenic and intravenous injection at multiple time points and immune environments, we will determine the cell states that confer selection for successful metastasis and adaptation to immune response. In Aim 3, we will investigate key regulators of metastatic phenotypes using multiplexed functional approaches. We will prioritize genes that are highly expressed or strongly induced in cell lines with enhanced metastatic ability and inactivate each candidate driver of metastasis in each cell line. By transplanting this pool of genetically altered cell lines, followed by Perturb-seq, we will determine the impact and transcriptional effects of each candidate driver on metastatic ability and organ tropism. Our study will uncover fundamental determinants of PDAC metastasis using comprehensive and systems-level approaches.

NIH Research Projects · FY 2025 · 2025-07

ABSTRACT SUMMARY Inhaled antibiotics are critical to the health of many patients, including those with cystic fibrosis (CF), a genetic disease characterized by thick, polymer-rich sputum and devastating chronic airway infections. Inhaled antibiotics are particularly important for suppression of Pseudomonas aeruginosa, a bacterial pathogen and major contributor to morbidity and mortality in CF. Unfortunately, the effectiveness of inhaled antibiotics is limited by poor antibiotic penetration of sputum biofilms – slimy communities of bacteria and polymers that colonize the airways of pwCF. We need improved approaches and therapeutic targets to enable diffusion of antibiotics through infected sputum. We have identified a microbial factor that prevents antibiotic diffusion in sputum. Pf bacteriophages (phages), are viruses produced by P. aeruginosa. Unlike most phages that lyse (kill) their bacterial hosts, Pf phages are produced without lysis. Instead, Pf phages function as structural elements in bacterial biofilms, including in CF airways. Pf phages organizes polymers present in both sputum and biofilms into a liquid crystalline state. The formation of these crystalline networks is driven by entropic, charge-based interactions between phages and polymers present in sputum. These biophysical assemblies prevent antibiotic diffusion and shield the bacteria within from antibiotic killing. We and others have reported that Pf phage is found in over 80% of adult CF patients. Moreover, Pf phage is associated with chronic P. aeruginosa lung infection, declines in pulmonary function, and resistance to several anti-Pseudomonal antibiotics. It may be possible to disrupt these crystalline structures by targeting Pf phage. Our preliminary data suggest that the antimicrobial peptide cathelicidin or LL-37, a component of the innate immune system, disrupts crystalline networks formed by Pf phage in vitro. Our model is that LL-37 and other cationic peptides bind to anionic Pf phage capsid proteins in ways that prevent liquid crystal formation and promote antibiotic diffusion. It may be possible to bundle conventional antibiotics with antimicrobial peptides to promote the diffusion of inhaled antibiotics in sputum. Unfortunately, however, the concentrations of LL-37 required to disrupt crystalline biofilms are toxic to human cells, Nonetheless, it may be possible to synthesize other cationic peptides that likewise promote antibiotic penetration but are non-toxic to cells. Here, we propose to identify peptides and small molecules that bind Pf, disrupt crystalline biofilms, and facilitate antibiotic diffusion in sputum. First, we will perform high-throughput screens to identify these molecules. Then, we will then evaluate these candidates in physiologically relevant and highly quantitative assays. Together, these aims represent a bold and radically unconventional approach to improving the efficacy of inhaled antibiotics against Pseudomonas airway infections in CF and other settings.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY/ABSTRACT Malaria in pregnancy is one of the leading causes of infant death globally. Placental malaria (PM) is the main mechanism by which malaria in pregnancy causes birth complications, such as preterm birth, stillbirth, and low birth weight. During PM, Plasmodium falciparum (Pf)-infected red blood cells sequester to syncytiotrophoblasts (STB) within the placental intervillous space, stimulating maternal immune cell recruitment and leading to other placental pathologic changes. With increasing gravidity, the severity of PM decreases and birth outcomes improve as pregnant women acquire Pf-specific antibodies (Abs) against a variant surface protein – VAR2CSA - after repeated Pf exposures. Previous work shows that neutralizing Abs against VAR2CSA are important for protection but do not entirely prevent neonatal complications, which suggests a key role for Ab-mediated effector functions. Work by our group and others demonstrate that Ab-mediated effector functions are crucial for naturally acquired immunity to malaria in children. We hypothesize that the myeloid compartment, phagocyte localization at the materno-fetal milieu, and Ab repertoire adapt with repeated Pf exposure in a gravidity-dependent manner, and that antibody-dependent phagocytosis is necessary for limiting PM pathogenesis and improving neonatal outcomes. To test this hypothesis, we will leverage an extraordinary biobank of peripheral and placental blood, plasma, and placental tissue collected from pregnant women enrolled in the DPSP clinical trial (U01 AI1431308). In Aim 1, we will determine how malaria and gravidity impact the circulating and placental blood myeloid compartment in pregnancy using single cell proteomic (CyTOF) and transcriptomic (scRNAseq) approaches. We will test whether Pf infections and increasing gravidity result in an enrichment of circulating, FcγR-expressing phagocytes. In Aim 2, we will determine how Pf parasitemia and gravidity shape myeloid cell - STB spatial relationships in placental tissue using novel spatial proteomic (MIBI-TOF) and transcriptomic (NanoString DSP) imaging approaches. We will test whether, with increasing gravidity, inflammatory maternal myeloid cell infiltration will decrease and maternal phagocytes will preferentially localize to the STB. Finally, in Aim 3, we will utilize in vitro models to determine how gravidity-induced Ab modifications influence Fc-mediated protection from malaria in pregnancy. Together, successful completion of these studies will inform vaccine development and therapeutic strategies to reduce the global burden of malaria in pregnancy.

NIH Research Projects · FY 2026 · 2025-07

ABSTRACT As evolving therapies enable a longer lifespan for patients with cancer, risk for kidney function decline, either pre-existing or developed as a complication of therapy, requires increasing attention. Many factors that render serum creatinine (sCr) or cystatin-C based kidney function assessments unreliable—including sarcopenia for the first or inflammation for the latter—are enriched in this population. Furthermore, widely used molecularly targeted cancer medications may inhibit tubular creatinine secretion and raise serum sCr; little is known about the long-term effects of these medications that once initiated, require long-term, daily intake. Therefore, uncertainty about the risk for clinically significant kidney function decline among patients on molecularly targeted cancer therapy is complicated by uncertainty about the performance of current kidney function measures in this population. Under mentorship from experts in care of patients with cancer and kidney disease, and in bioinformatics, Dr. Ziolkowski will address critical data gaps in assessment of kidney function among patients with cancer. She will take coursework in bioinformatics and apply machine learning methodology to harness the rich health data available for a person with cancer. In a retrospective cohort of patients treated at Stanford Health care, our team will determine the two-year incidence rate and risk factors for progressive chronic kidney disease (CKD) for patients on cancer therapies associated with an acute rise in sCr at drug initiation, with the hypothesis that long term exposure to these medications may contribute to progressive CKD. We will obtain sCr, cystatin C, and kidney biomarkers of tubular injury pre- and post- drug start to characterize the acute effects of drug initiation. We will develop a risk assessment model using machine learning for patients on molecularly-targeted cancer therapies to predict risk of progressive CKD, with the hypothesis that integrating multi-dimensional individual-level data in the electronic medical record will have greater prognostic yield for progressive CKD than estimated glomerular filtration rate cut points. We anticipate incorporation of computed tomography (CT) measures of muscle mass and kidney size improve our risk stratification. We will externally validate the prediction model at Mount Sinai Health System. A risk stratification model could identify patients at risk for progressive CKD, providing opportunities for appropriate counseling, drug dosing, avoidance of nephrotoxins, and treatment with renal-protective medications while those at lower risk could have expanded treatment options. Additional training in risk assessment methodologies and machine learning will enable Dr Ziolkowski to become an independent investigator, applying the latest bioinformatics methodologies to improve kidney function outcomes among patients with cancer. She will apply for an investigator led (R01) grant to determine clinical utility of our risk assessment model, using simulated and real-world prospective data, and broaden the risk assessment methodology to other cancer populations.

NSF Awards · FY 2025 · 2025-07

Earth’s lands, waters, and ecosystems underpin human well-being and economic prosperity, providing a wide array of ecosystem services from food to clean water to protection from flooding. An active ecosystem service research community continues to deepen the understanding of where and how nature contributes to human well-being, alongside growing demand from governments, financial institutions, companies, and society in support of decision-making and environmental management. This project will scope and develop a plan for transitioning one of the most widely used tools for mapping and quantifying ecosystem services into a full open-source ecosystem, accelerating innovation in and the use of ecosystem services science across the public and private sectors. By engaging a broad set of users and contributors, this project will support increased partnerships among research, industry, and policy makers to develop, maintain, and use the software and the ecosystem services information it provides, for the benefit of people and nature. This project will scope the evolution of the open-source InVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) suite of ecosystem service models into an open-source ecosystem with an open and engaged community of users and contributors. The project team will undertake surveys to understand the existing user and contributor community, develop guidance and onboarding materials to increase the intellectual contributor community, and organize a workshop of InVEST contributors and developers to create a plan for evolving InVEST governance and for measuring success. Through these efforts, the project team will chart a path towards an open-source ecosystem for InVEST that makes it easier and faster to incorporate ecosystem service science into research and decision-making. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY: Schwannomas are the most common sporadic nerve sheath tumor which arise from the Schwann cells that ensheathe and insulate axons of the peripheral nervous system. Schwannomas can cause significant morbidity depending on location and adjacent structures that are compressed over time, particularly those involving cranial nerves such as hearing loss for vestibular nerve tumors. Complete resection can be curative, but may result in permanent nerve injury. Furthermore, resection is not always possible depending on tumor location that can result in recurrence and progression over time, necessitating additional surgery or radiation therapy for disease control and symptom management. Thus, there is a pressing clinical need for more efficacious treatment options. While studying schwannomas arising in the setting of known predisposition syndromes due to germline mutations involving NF2, LZTR1, and SMARCB1 has shed substantial light on schwannoma pathogenesis, the molecular drivers responsible for sporadic schwannomas have not been fully resolved to date. Through genomic profiling studies, we discovered that 22/77 (29%) of sporadic schwannomas lack alterations in known schwannoma-associated genes and instead harbor recurrent in-frame insertions or deletions in SOX10, which encodes a transcription factor responsible for controlling Schwann cell differentiation. These insertions/deletions are uniformly somatic (tumor-acquired) and localize at the C-terminal end of the HMG box domain responsible for DNA binding. Our preliminary analysis has revealed that schwannomas harboring SOX10 indel mutations often occur along paraspinal and non-vestibular cranial nerves and have high local recurrence rates. Based on our identification of recurrent SOX10 indel mutations within the DNA binding domain, we hypothesize these mutations are likely to drive schwannoma development through altered chromatin binding and transcriptional reprogramming that blocks terminal differentiation of immature Schwann cells resulting in a persistent progenitor-like state. To test this hypothesis, we propose three aims utilizing our unique cohort of clinically annotated and molecularly-genotyped human schwannoma tumor specimens, primary human schwannoma cultures and Schwann cell models with wildtype or tumor- derived mutant SOX10 alleles, and a conditional Sox10flox-K172_Y173dup knock-in mutant mouse strain we have generated. In Aim #1, we will perform multi-omic profiling to determine whether SOX10 mutation defines a unique subtype of schwannoma with distinct histopathologic, transcriptomic, epigenomic, and cellular features. In Aim #2, we will evaluate whether tumor-derived SOX10 mutations result in altered chromatin binding and transcriptional reprogramming. In Aim #3, we will investigate whether SOX10 indel mutations are sufficient and necessary to drive schwannoma growth and explore potential therapeutic vulnerabilities of SOX10 mutant schwannomas. Together, these studies aim to reveal fundamental properties of Schwann cell differentiation and pinpoint the tumorigenic mechanism of a substantial fraction of clinically aggressive schwannomas.

NIH Research Projects · FY 2025 · 2025-07

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains the only curative therapy for many malignant and non-malignant diseases. Despite advances in allo-HSCT, graft-versus-host disease (GVHD) remains a significant cause of morbidity and mortality in patients. GVHD is marked by the activation of donor alloreactive T cells which target and damage healthy recipient tissues. This project’s central hypothesis is that donor alloreactive T cells differentiate into immunoregulatory or pro-inflammatory effectors depending on the environment into which they are transplanted and that identifying these alloreactive cells by high-resolution sequencing can predict clinical GVHD. My long-term career goal is to use high-resolution bioinformatics to dissect the distinct differentiation pathways which lead to tolerance versus alloreactivity after adoptive cell therapy. Consequently, this proposal has three main aims. During the mentored K99 phase of this award, we first hypothesize that alloreactive T cell subsets develop their phenotype after allo-HSCT and that the detection of anti-human T cells can predict GVHD after allo-HSCT. Second, during the independent R00 phase, we posit that these and other T cell subsets arise from naïve T cells in response to allo-antigen and we will investigate the mechanisms which drive this differentiation in both human and mouse. Third, we propose that specific T cells which mediate alloreactive GVHD responses will be identifiable after allo-HSCT and experiments in Aim 3 will utilize high-resolution sequencing techniques to identify and track alloreactive T cells in both xenogeneic and murine models of GVHD. These scientific aims will be explored alongside additional training in human experimental methods, advanced bioinformatics and machine learning, career development, and clinical translation with the support of a highly experienced, multidisciplinary mentoring team. The overall objective of this proposal will be to develop the ability to track alloreactive T cells in patients to predict disease onset prior to the emergence of deleterious symptoms. These novel results will be directly applicable to the field of allo-HSCT as well as other contexts involving alloreactive T cell responses including solid organ transplantation, chimeric antigen receptor T cell therapy, and other adoptive T cell therapies.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY Our understanding of fundamental biological processes is both driven and limited by our ability to visualize macromolecular structure and organization. Access to rapid characterization of proteins and nucleic acids at the single-molecule level is now essential for scientists to push the frontiers of biological research and is critical for the application of this knowledge for the improvement of human health. This proposal requests funds to purchase the TwoMP Mass Photometer with MassFluidix Microfluidic Technology (Refeyn, Inc.) for the analysis of biological molecules in the size range from kilodalton to megadalton. The fast imaging and high-resolution capabilities of the TwoMP are also designed to overcome the speed and sample abundance limitations of single-molecule imaging. We have identified the TwoMP as the sole commercial product which meets our requirements, is state-of-the-art and cost-effective. This instrument will be a shared resource in a well-established core facility at Stanford, the Macromolecular Structure Knowledge Center (MSKC). The TwoMP mass photometer will support more than twenty cutting edge health care NIH projects described in this proposal. These projects cover a wide range of topics, including: energy metabolism (Long); drug discovery (Gray); cholesterol (Welander); gene therapy (Qi); insulin-independent pathways (Svensson); neurotransmission (Maduke); biomaterials (Mai), brain development (Fame); signaling (Jackson); peptide biomimicry (Barron); receptors (Barnes); RNA (Das); transcription factors (Khavari); oxidative damage (Wang); allergy (Jardetzky); mRNA (Martinez); genome assembly (Straight); autoantibodies (Lanz); nanoparticles (Sharaf); chromatin architecture (Altemose); celiac disease (Khosla); pandemic viruses (Lin); and vaccine development (Kim). These studies address critical functional and structural questions about fundamental biological processes and span NIH research areas with implications for diverse aspects of human health and disease, including diabetes, cancer, dementia, obesity, Alzheimer's disease, influenza, and pandemic viruses. The speed and low sample requirements of the TwoMP instrument are critical for all these projects, where sample scarcity and instability are often a significant challenge. These combined capabilities will drastically enhance our NIH funded faculty ability to conduct biophysical research by providing a shared resource that meets their biomolecular analysis needs.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY Head and neck squamous cell carcinoma (HNSCC) is a leading cause of cancer mortality and treatment can cause significant functional impairments for survivors. Despite advances in treatment, survival rates have not meaningfully improved across many disease sites over the last several decades. Cancer metabolism is known to play important roles in tumor progression and treatment resistance, and HNSCC tumors undergo metabolic changes to increase energy production through glycolysis; however, the underlying mechanisms for this metabolic shift are poorly understood and targeting cancer metabolism remains elusive. The broad research goal of this proposal is to define how ACTL6A, a gene commonly amplified in HNSCC and a component of the SWI/SNF chromatin remodeling complex, contributes to metabolic reprogramming in HNSCC. Preliminary work shows that ACTL6A decreases mitochondrial potential of HNSCC cells and may be necessary for maintenance of a cancer stem cell like population of cells with low mitochondrial potential. These findings suggest ACTL6A expression levels may be important for regulating rates of ATP production from oxidative phosphorylation and glycolysis in cancer cells. To uncover how ACTL6A regulates cell metabolism, Aim 1 will quantify energy production in settings of low and high ACTL6A levels using HNSCC cells. An additional sub-aim will use an in vivo tumor model with a cancer drug targeting aerobic respiration along with a drug targeting SWI/SNF to establish a new strategy for treating HNSCC. Aim 2 will use CUT&RUN to define SWI/SNF distribution of a cell population with low mitochondrial activity and increased cancer stem cell gene expression whose population is nearly completely attenuated with ACTL6A knockdown. Together, these aims will establish the role of ACTL6A on cell metabolism, offer new combinations of therapies that may be effective against HNSCC, and open a novel area of research into regulation of cancer stem cell metabolism by SWI/SNF.

NIH Research Projects · FY 2025 · 2025-07

PROJECT SUMMARY Early supportive caregiving relationships can mitigate the potential negative effects of early adversity. As adverse childhood experiences and their associated risks represent a significant public health crisis, understanding how to promote positive early relationships is crucial. Early childhood presents a window for intervention in the caregiver-child relationship, with the potential to leave a lasting impact on future outcomes. Investigating features of dynamic caregiver-child relationships in naturalistic settings provides valuable insights into the mechanisms that can promote positive developmental outcomes even in challenging conditions. A key feature of successful dyadic interactions is behavioral synchrony, which involves the dynamic coupling and coordination of temporally linked behavior between social partners. Synchronous interactions shape children's self-regulation, school adjustment, and brain development. However, while early interventions target the caregiver-child dyad, current research often fails to measure changes in dyadic mechanisms. Early caregiving intervention programs offer an ideal opportunity to study dynamic changes in behavior synchrony. This project proposes to develop innovative global ratings and coding measures for behavioral synchrony, to examine whether these dyadic mechanisms are sensitive to change across early intervention. This approach addresses a significant gap in understanding how dyadic interactions adapt within early intervention contexts, at a global and microsocial level. Exploring this question mechanistically will enhance our understanding of how and why specific program components are associated with desired outcomes. The current project uses a collection of recorded caregiver-child interactions from a scalable, conceptually-based, and innovative early video coaching intervention in families experiencing economic adversity. Aim 1 proposes the development of a global behavioral synchrony rating scale to code naturalistic caregiver-child interactions, to examine synchrony as a dyadic mechanism. Aim 2 leverages the time series nature of the dyadic data in video clips of caregiver-child interactions. It introduces a micro-level coding system for behavioral synchrony to assess dyadic behaviors in detail, second by second. Aim 3 will investigate whether the proposed measures of behavioral synchrony are sensitive to change in the intervention group compared to an active control group. These contributions will meaningfully add to the extant theoretical, conceptual, and translational work by innovatively examining behavioral synchrony. Guided by Sponsor Philip Fisher at Stanford University, this training and research plan will advance my journey toward becoming an independent researcher and future tenure-track professor. My research focuses on how children's dynamic interactions with their environment enhance well-being and healthy development.

NIH Research Projects · FY 2025 · 2025-07

Project Abstract Our project aims to harness the transformative potential of Foundation Models (FMs) in the fields of clinical neuroscience and psychiatry. In recent years, a successful new paradigm for building artificial intelligence systems has emerged: train a single model on a vast amount of heterogeneous, unlabeled data and adapt it to various applications. While such FMs have demonstrated unprecedented capabilities in fields as diverse as natural language processing and medicine, their potential for decoding the complexities of human functional brain imaging data remains largely untapped. Crucially, conventional data analysis methods for functional brain imaging (fMRI) are hampered by the need for large, labeled datasets, their inability to capture the intricate spatiotemporal dynamics associated with psychopathology, and the recurrent issues of heterogeneous data and class imbalance in clinical brain imaging studies. To address these challenges, we propose to develop an integrated framework that combines FMs with Explainable Artificial Intelligence (XAI) techniques. This framework will be used to analyze large open-source human brain imaging datasets and phenotypic data across multiple psychiatric disorders. Moreover, our innovative FMs will provide a robust framework for effectively integrating heterogeneous datasets and addressing the recurrent issue of class imbalance in clinical brain imaging studies. Building on our highly promising preliminary results, our team, working with the Stanford Center for Foundation Model Research – a leading institute dedicated to FM research – is uniquely positioned to achieve the following research objectives: First, we will develop FMs specifically designed for large, open- source task-free fMRI datasets such as the HCP, Lifespan Connectome, Human Connectomes Related to Human Disease, and ABCD, among others. Second, we will fine-tune FMs to investigate multiple highly prevalent psychiatric disorders and predict clinical symptom severity. Lastly, we will identify personalized neurobiological features and associated brain networks underlying these disorders. Our FMs will obviate the need for large, labeled datasets, exhibit robustness against class imbalance, and generalize to heterogeneous data. Our approach is theoretically grounded, building upon unifying models of psychopathology that emphasize the role of key control brain networks in multiple psychiatric disorders. By pioneering state-of-the-art FMs tailored for functional brain imaging, our work aims to equip researchers with a transformative toolset for investigating the neurobiology of a range of psychiatric disorders within a unified framework. This will lay the groundwork for more precise diagnostic and treatment strategies in psychiatry. Moreover, the open sharing of FMs and associated data analysis code will catalyze future research, significantly amplifying the impact and scope of our work.

NIH Research Projects · FY 2025 · 2025-07

Reinvigorating the overdominant model of tumor suppression Tumor suppressor genes (TSGs) play essential roles in cancer pathogenesis, from cancer initiation to progression and therapy responses. More than 50 years after Knudson proposed the “two-hit” model of tumor suppression, this paradigm remains fundamental to the field of cancer biology and was later complemented by the “one-hit” or haploinsufficiency model of tumor suppression. In contrast, a third mode of tumor suppression posits a non-monotonic relationship between TSG dosage and tumor fitness, where inactivation of one gene copy is beneficial, but the loss of both is less beneficial or even detrimental—a phenomenon due to genetic overdominance. Utilizing CRISPR/Cas9-based somatic genome editing and tumor barcode sequencing (Tuba- seqUltra), combined with gRNA cutting efficiency information, I identified seven candidate overdominant TSGs in the oncogenic KRAS-driven lung adenocarcinoma, notably including Suz12 and Eed from the polycomb repressive complex 2 (PRC2). In this proposed study, I hypothesize that overdominance is a significant factor shaping the functionality of many TSGs and that overdominant TSGs are enriched in chromatin regulators. First, I will refine the Tuba-seqUltra platform to systematically measure the overdominant effect of TSGs in vivo and explore the interplay between different oncogenic backgrounds and overdominant tumor suppression (Aim 1). To gain insight into the molecular mechanism of overdominant tumor suppression, I will investigate the phenotypic consequences of heterozygous and homozygous deletion of the key PRC2 members Eed using single-cell RNA-Seq and single-cell ATAC-Seq (Aim 2). Finally, to study the extent of overdominant tumor suppression in lung adenocarcinoma, I will use machine learning methods to quantify the overdominance of 200 known and putative TSG. This will determine the prevalence of overdominance in tumor suppression and assess whether overdominant TSGs are enriched in dosage-sensitive pathways like those involving chromatin regulators (Aim 3). This study will establish robust methodologies for researching overdominant TSGs in cancer, evaluating their prevalence, and deciphering the underlying molecular mechanisms. This could profoundly impact our understanding of cancer biology and influence future therapeutic strategies.