Stanford University

NIH Research Projects · FY 2025 · 2024-02

PROJECT SUMMARY: Most plants have evolved the capacity to synthetize large, complex metabolites that potently target animal physiology. While we have adopted some of these molecules as useful medicines (Taxol, colchicine, morphine, Velban), we understand a miniscule fraction of the chemistry that most plants are capable of. Most of the 3k-15k enzymes per genome involved in specialized metabolism have unknown roles, including in the genomes of plants we regularly consume as food. Learning the roles of these genes would not only illuminate what molecules a plant can synthesize, but also would allow us to (1) produce these molecules exogenously, (2) to directly assess their bioactivity, (3) make informed dietary decisions, and (4) genetically delete molecules from dietary plants. However, this is a challenging task because plant genetics is limited, plant genomes are large, and biosynthetic genes are often silent except under specific conditions. The aim of this proposal is to develop and apply tools that will systematically activate and identify different biosynthetic gene sets, efficiently defining a plants’ arsenal of biochemical pathways. In Aim 1, I will achieve this by the targeted activation of hundreds of transcription factors that regulate diverse biosynthetic pathways. During my K99 phase, I will develop this approach in tomato, a plant with relatively good genetic tools, optimizing a combination of multiplexed gene activation and single-cell measurement. I will also train with Dr. Jennifer Brophy’s lab in in vitro plant cell assays, so that during the R00 phase I can expand this approach to studying the biosynthetic dark matter of wheat and sorghum genomes. In Aim 2, I develop a variant of this methodology compatible with genetically intractable plants. I will use panels of chemical perturbations to activate biosynthesis, and a newly developed microfluidic technology to isolate and characterize cells with specialized biosynthetic roles. This will enable an accelerated dissection of the biochemistry of medicinal plants like Taxus, our sole source of the blockbuster drug Taxol. To achieve my long-term goal of becoming an independent researcher and pioneering a new scale of plant science, I have assembled a mentorship team that will complement my training in systems and synthetic biology: Dr. Elizabeth Sattely (mentor, plant biochemistry), Dr. Polly Fordyce (co-mentor, microfluidics and high throughput biology), Dr. Jonathan Weissman (advisor, single-cell biology and screening), Dr. Jennifer Brophy (advisor, plant synthetic biology), Dr. Bo Wang (advisor, single-cell technologies in noncanonical organisms). My mentors and I have designed a plan with specific goals for scientific training, career development coaching, scientific presentation, and coursework. The resources and thriving scientific environment at Stanford will help me achieve these goals. With the training and mentorship provided by this K99 opportunity, I anticipate a smooth transition to managing my own research group. There, I will establish a unique research program to develop systematic approaches to study the biosynthesis and health impacts of specialized metabolites.

NIH Research Projects · FY 2026 · 2024-02

PROJECT SUMMARY A cancer diagnosis and its subsequent treatment affects whole patient health -- disrupting the full spectrum of physical, social, emotional, and functional quality of life. An estimated 83% of cancer patients report low to very low quality of life. Individuals receiving chemotherapy report experiencing over 13 concurrent symptoms, including fatigue, sleep difficulties, and pain. 30%-40% of people with cancer report significant psychological symptoms, including anxiety and depression. Such disruptions in mental health and quality of life, in turn, exacerbate physical symptoms and can worsen clinical outcomes. Despite growing evidence of the complex, interconnected pathways linking the mind and body, scalable interventions that efficiently target whole-person health upon diagnosis of a life-altering disease such as cancer have not yet been developed. Existing medical treatments, focus on the physiological aspects of the disease. Existing psychological treatments, such as Cognitive Behavior Therapy (CBT) and Mindfulness Based Stress Reduction (MBSR), apply a broad range of cognitive and behavioral strategies to reduce diffuse symptoms of depression and anxiety. Such ancillary programs for psychological care are routinely proposed as methods to reduce distress and symptoms, restore function, and improve quality of life, but they are frequently inaccessible to patients and notoriously difficult to scale. The MINDSET intervention aims to close this major treatment gap and promote whole patient health by targeting patient mindsets at the point of diagnosis. This proposal builds on our extensive foundational research completed over the past 6 years with support of the NIH New Innovator Award (DP2 AT009511) to test a novel mHealth intervention targeting patient mindsets at the point of diagnosis. In our prior research, we showed that a 2.5-hour digital MINDSET intervention significantly improves whole patient health (physical, social, and emotional functioning as measured by the FACT-G) in patients undergoing systemic treatment for cancer with curative intent compared to a Treatment as Usual (TAU) control. Leveraging our interdisciplinary team of experts in oncology, psychology, psychiatry, mHealth, and biostatistics, we aim to expand on this prior research to address this large and costly gap in clinical care. We propose a fully decentralized Phase 3 randomized controlled trial in which 440 cancer patients treated for non-metastatic solid tumors and hematological malignancies will be allocated to either a (1) MINDSET or (2) Matched Attention Control (MAC) condition. Patient-reported primary outcomes (FACT-G Total Score) and secondary outcomes (anxiety, depression, affect, sleep, coping, symptom distress, patient activation/engagement, and inflammatory biology) will be assessed at weeks 0 (baseline), 2, 4, 6, and 10. Durability of the effect of the intervention will be measured at two follow-up timepoints: 3 months, and 6 months after study completion. By promoting health at the psychological, behavioral, and biological levels, MINDSET interventions have the potential to become a highly impactful and complementary tool for promoting whole patient health.

NIH Research Projects · FY 2026 · 2024-02

PROJECT SUMMARY/ ABSTRACT Hematopoietic stem cells (HSCs) lie at the top of the blood hierarchy and are capable of giving rise to all blood cells of an organism. Consequently, their use has enormous therapeutic potential for the treatment of blood diseases, and generation of HSCs in vitro is a central aim in regenerative biology. Despite this clinical need, we lack protocols that allow us to efficiently generate HSCs in vitro that are capable of long-term engraftment and multi-lineage output. A major hindrance is our incomplete understanding of how HSCs are made in vivo. For instance, although it is established that blood cells develop from endothelial cells in multiple sites throughout the embryo, we still do not know which embryonic sites produce long-term HSCs, nor how site of origin impacts on life-long stem cell function or behavior. Furthermore, we are limited in our understanding of the intrinsic and extrinsic cues driving functional heterogeneity in hemogenic endothelial cells. This project proposes to use powerful next-generation barcoding technology to enrich our understanding of the embryonic origin of HSCs and the hemogenic endothelial cell states that give rise to blood to allow us to harness the process in vitro. Dr. Bowling conducted her graduate work in developmental biology and during her postdoctoral training has focused on the development of next-generation DNA barcoding tools for performing single cell, inducible cell lineage tracing in vivo. Equipped with this skillset, Dr. Bowling plans to use innovative cellular barcoding techniques to, for the first time, catalog the precise endothelial origins of long-lived blood progenitors in the mammalian embryo (Aim 1). Furthermore, she will perform in-depth characterization of the heterogeneous endothelial cell states that give rise to distinct blood cells in the embryo (Aim 2). The knowledge generated from these experiments have the potential to answer major, long-standing questions in the field of developmental hematopoiesis and transform our basic understanding of the steps leading to blood generation, and therefore revolutionize protocols for HSC generation in vitro. Dr. Bowling is supported by a panel of mentors and consultants who are world-class researchers in hematology, developmental biology, and technology development. Her mentors Drs. Fernando Camargo and Leonard Zon have made exceptional contributions to the field of hematopoiesis and also have outstanding track- records for mentorship. Dr. Bowling will gain further scientific training and career development support from her scientific committee: Drs Jay Shendure, Trista North and Berthold Gottgens. Finally, she will benefit from carrying out her research program in the scientifically stimulating and resource-rich environment of Boston Children’s Hospital and Harvard Medical School. The aims in this proposal will allow Dr. Bowling to build on her skills to gain expert knowledge in the computational analysis of sequencing datasets and the use of induced pluripotent stem cells, in preparation for her transition to independence. As a result, she will establish a unique niche for resolving important, clinically-relevant questions in hematopoietic development as an independent researcher.

NIH Research Projects · FY 2026 · 2024-02

The glycans present on IgG antibodies are key determinants of immune complex (IC) signaling by FcγRs, with modulations in IgG glycosylation triggering distinct antibody effector functions in response to infection or immunization. While FcγR signaling pathways are broadly characterized, with established functions in cell maturation, phagocytosis, production or release of soluble factors, and suppression or modulation of activating signals, little is known about how these responses are integrated at the tissue level. Of particular interest is understanding how IgG glycoforms regulate antibody-mediated functions in the lung, the immune interface for respiratory pathogens and a tissue with unique cellular features, including lung-specific, FcγR-expressing cells. Using a newly developed system in mice that express human FcγRs, we have found that IC glycosylation, specifically fucosylation and sialylation, can regulate lung inflammatory responses. We now propose to examine the role of distinct IgG IC glycoforms in regulating the outcomes of SARS-CoV-2 and IAV virus infections, as we have observed that different patterns of IC glycosylation correlate with disease risk in these two infections. We will study how signaling by differentially glycoslylated IC impacts viral and host immune readouts in SARS-CoV- 2 and IAV virus infections, and we will define molecular mechanisms governing the regulation of lung inflammation by IC. Our hope is that these studies will reveal aspects of how the host IgG glycome contributes to heterogeneous outcomes of infection with SARS-CoV-2 and IAV viruses and reveal mechanisms that can be targeted to prevent the severe lung inflammation that leads to ARDS and mortality in these and potentially other respiratory infections.

NIH Research Projects · FY 2026 · 2024-02

PROJECT SUMMARY This proposal addresses the ethical, social, and policy implications of an emerging, controversial field of research: social and behavioral genomics (SBG). Using molecular, genome-wide data, SBG examines whether and how genetic differences between individuals shape differences in traits and outcomes such as educational attainment and math ability. Today, SBG is more accessible than ever. Members of the public can easily acquire direct-to-consumer genetic tests for intelligence, math ability, and sexual promiscuity, among others. Despite the growing availability of SBG data, few policies and incentives exist to consider SBG’s downstream implications. Further, there remains a critical lack of breadth regarding who gets to define the harms and benefits of SBG. As such, SBG engenders a host of difficult ELSI questions to be addressed by this K01: What are the downstream implications of social and behavioral genomics as identified by a broader set of stakeholders than ever before (i.e., SBG researchers, journal editors, journalists, members of industry, parents, and educators)? What are stakeholders’ roles in producing, promoting, and/or mitigating the downstream harms and benefits of SBG? How might the potential harms of SBG be mitigated against and its potential benefits promoted? To address these questions, this proposal includes a plan of research that will achieve three aims. Aim 1 will use in-depth, semi-structured interviews to investigate stakeholders’ (i.e., SBG researchers, journal editors, journalists, members of industry, parents, and educators) roles in the conduct and translation of SBG and their perspectives on the downstream harms/benefits of SBG. Aim 2 will adapt The Ethical Matrix to determine stakeholders’ roles in producing, mitigating, and/or promoting SBG’s downstream harms and benefits for the purpose of locating social responsibility. Finally, Aim 3 will draw upon deliberative engagement theory and methods to design, implement, and evaluate a series of participatory sessions that bring a selection of stakeholders interviewed in Aim 1 together to: (1) discuss (dis)agreements about Aim 1 and 2 findings; and (2) identify strategies for mitigating the harms and promoting the benefits of SBG. Building on Dr. Martschenko’s strong background in mixed-methods education research, policy development, and the ethical, legal, and social implications (ELSI) of human genetics/genomics, these aims will be achieved with a career development plan that includes required training and coursework in science, technology, and society (STS) studies and deliberative engagement theory and methods. Dr. Martschenko’s career development plan will also strengthen her interdisciplinary ELSI networks. The career development plan is supported by a team of esteemed, interdisciplinary scholars at Stanford University, The Hastings Center, the University of California, San Francisco, and the University of California, Los Angeles: Mildred K. Cho (primary mentor), Erik Parens (co-mentor), Janet Shim (co-mentor), Julie Harris-Wai (co-mentor), Barbara Koenig (advisory committee), Aaron Panofsky (advisory committee), and Ben Domingue (advisory committee).

NIH Research Projects · FY 2026 · 2024-02

I am a physician-scientist dedicated to caring for people with schizophrenia (SZ) and studying the pathophysiology of SZ to identify disease-modifying therapeutic interventions. Converging evidence in the field points to abnormalities in synapses, and perhaps a reduction in specific synaptic subtypes in distinct cortical layers due to ongoing synapse elimination, that may result in abnormal neuronal signaling that interfere with cognition. However, methodological limitations have impeded progress in testing the neuroanatomical basis of this hypothesis. In this project, a novel multiplex imaging method, Array Tomography, will be applied to a large cohort of dorsolateral prefrontal cortex (DLPFC) sections of postmortem brain samples from individuals with SZ and controls in combination with transcriptomic data to test whether individuals with SZ a have a lower synaptic density of excitatory and inhibitory synapses in layer 3 (L3) of the DLPFC that correlate with high C4A gene expression. Specific aims include: [1] Comparing both the synaptic density of excitatory and inhibitory synapses (and ratios of excitatory to inhibitory synapses) in DLPFC L3 from 55 SZ and 55 control donors, to test whether the synaptic density of either (or both) primary synaptic subtypes differs between SZ and controls, while also testing whether an imbalance in excitatory and inhibitory synaptic density could be the basis of the imbalance in excitatory and inhibitory signals that may be disrupting cognition (E/I imbalance). [2] determining the specificity of the synaptic density abnormalities in L3 by measuring the excitatory and inhibitory synaptic density in the same samples in layer 5, and [3] calculating the correlation between excitatory and inhibitory synaptic density in DLPFC L3 and C4A gene expression, as measured by bulk RNA- sequencing [RNA-seq]. Demonstrating a strong correlation between high C4A gene expression and low synaptic density would provide supporting evidence for the excessive synaptic pruning hypothesis and C4A gene expression as a primary mediator of synaptic density. Work on this project will provide empirical data for leading hypotheses in the field while supporting my Career Development. I will gain extensive experience in using Array Tomography to study detailed synaptic architecture and work towards identifying critical factors that regulate synapses in SZ by coupling with transcriptomic data. This experience will build on my prior experience so I can work on the underlying synaptic pathology in SZ using a combination of postmortem tissue and in vivo model systems as an independent investigator. Goals of this Career Development Award will be overseen by my primary mentor, Professor Urban, expert in -omis analyses in psychiatric disorders. An outstanding Mentoring Team at Stanford University will support my career development in gaining experience in using Array Tomography, bioinformatic analyses, deepening my knowledge of synaptic biology, while building the publication record and preliminary data for a successful transition to an independence.

NIH Research Projects · FY 2026 · 2024-02

SUMMARY Chimeric antigen receptor (CAR) T cell therapies represent a major advancement in the cancer immunotherapy (IOT) field and have shown remarkable success particularly for treating hematologic malignancies. Despite their meteoric rise, these “living drugs” face a number of challenges that have limited their widespread utility (especially for solid tumors), including the inability to discern cell fate after infusion. Growing investment in developing new CAR T cell therapies and the desire to optimize existing ones highlights the urgent unmet need for techniques that permit non-invasive, longitudinal monitoring of CAR T cell fate. An imaging method that enables sensitive detection of activated T cells, including CAR T cells, in vivo has the potential to reveal mechanisms underlying failures, thus enhancing our ability to design and optimize IOTs. Immuno-positron emission tomography (immunoPET), a rapidly expanding area of molecular imaging that combines ultra-specific antibodies or antibody fragments with PET radioisotopes, has enormous potential to shed light on the in vivo spatiotemporal distribution and functional status of CAR T cells. Unfortunately, there are no clinically approved gold-standard immunoPET radiotracers, or other imaging techniques, that target T cell activation markers. Recently, we identified OX40 and ICOS (two cell-surface antigens that are highly and specifically upregulated on activated T cells) as promising biomarkers for imaging T cell responses. Subsequently, we developed the first PET radio tracer s to image murine OX40 and ICOS by radiolabeling monoclonal antibodies (mAbs) with 64Cu or 89Zr and demonstrated their utility for quantifying and tracking activated T cells in a variety of cancer IOT models. We also verified that both OX40 and ICOS are significantly upregulated on activated human and murine CD19- CAR T cells, and that ICOS immunoPET enables visualization of the latter in vivo. In this R01, we propose to build on our promising preclinical data by developing the first human-specific radiotracers for OX40 and ICOS, utilizing mAbs and F(ab’)2 fragments. We have performed preliminary 89Zr- radiolabeling of human-OX40 and - ICOS mAbs and demonstrated their specificity for human activated T cells. Additionally, we have acquired highly encouraging in vivo imaging and stability data for our novel human OX40-mAb radiotracer. Aim 1 involves optimizing radiolabeling for ICOS and OX40 mAbs, generating and labeling F(ab’)2 fragments, in addition to thoroughly assessing stability, affinity, and immunoreactivity of all four new radiotracers. We will then investigate potential biological effects and in vivo binding of these radiotracers to activated human T cells (Aim 2) and to CAR T cells, the latter engineered to target liquid and solid tumors (Aim 3). Finally, we will assess the safety of these radiotracers for CAR T cell imaging (Aim 3). Completing these experiments will provide invaluable data on the sensitivity and specificity, as well potential biologic effects, of all four imaging agents, guiding their optimization for our long-term goal of clinical translation. Such OX40 and ICOS PET radiotracers will afford a new approach for monitoring CAR T cells and other IOTs with great potential to increase their clinical success.

NIH Research Projects · FY 2026 · 2024-02

SUMMARY While single-cell methods for analyzing gene expression are becoming a standard tool for unpacking cellular heterogeneity and understanding complex tissues in health and disease, other molecular features, especially open chromatin landscapes via ATAC-seq, but also surface protein abundance and the presence of CRISPR guides, are rapidly expanding in their application. Indeed, commercial platforms for generating diverse single- cell data sets have led to an immense increase in scale of these data, and methods for split-and-pool based assays and decreasing sequencing cost all presage an exponentially increasing corpus of future large-scale datasets. We developed ArchR, an analysis infrastructure specifically designed for analysis of large-scale single- cell (sc) ATAC-seq data sets that enables a diverse suite of complex analysis (including QC, doublet removal, iterative TF-IDF clustering, approximation methods for large-scale data sets, trajectory analysis, RNA-seq integration, track visualization, marker peak identification, etc.), all with minimal computing hardware requirements. We estimate that ArchR has thousands of active users and is rapidly becoming the “go to” analysis software for large scATAC-seq data sets. To further extend the utility of ArchR for analyzing multi-omic data sets, we will first engineer substantial improvements to computational efficiency of underlying single-cell computational infrastructure. To do this, we will (1) encode our fundamental matrix operations in C++ to enable streaming data matrix access, thus reducing memory requirements and effectively “lifting the cap” on the number of cells capable of being analyzed through rapid on-the-fly calculations of diverse operations and (2) implement and benchmark efficient on-disk storage using bitpacking algorithms. These data structures and atomic operation libraries will be shared with the genomics community (and are being integrated into the popular Seurat package), allowing repurposing of these performance improvements. Second, we will develop, implement, and benchmark powerful analytical tools for the analysis of large, diverse, and/or multi-omic datasets. We will enable the handling of diverse independent and simultaneously acquired (multi-omic) data types including RNA-seq, ATAC-seq, ADT (CITE-seq), and CRISPR-based perturbation methods. We will develop accurate methods for cross-manifold data linkage for distinct data sets, forced-projection and regression analysis, multi-modality cell clustering, joint analysis of single-cell molecular data sets with CRISPR-based perturbations, single-cell inference of enhancer function via correlation and the “ABC” model, and identification of continuous differentiation trajectories and chromatin “potential.” Finally, we will develop plug-and-play cell type specific deep learning models for prediction of the regulatory effects of noncoding sequence changes. These models will learn single-cell chromatin accessibility profiles from DNA sequence to predict the cell type-specific effects of noncoding sequence changes. We will create a user-friendly system for training, deployment, and sharing sequence-based models of cell type- specific chromatin accessibility, bringing cutting-edge machine learning for functional genomics to wide use.

NIH Research Projects · FY 2026 · 2024-02

Rectal cancer is a common cancer and leading cause of cancer death. The majority of rectal cancer patients are diagnosed with stage II or III locally advanced disease. Neoadjuvant chemoradiotherapy followed by total mesorectal excision is the standard of care for treating these patients. However, the response to neoadjuvant therapy is highly heterogeneous. While approximately 20% patients achieve a pathologic complete response and long-term disease control, over 30% patients will develop distant metastasis despite multimodality treatment. Because total mesorectal excision is associated with significant morbidity and poor quality of life, organ preservation (i.e., watchful waiting without surgery) is recommended for patients who are deemed to achieve a complete response. On the other hand, for patients at high risk of recurrence, total neoadjuvant therapy with induction chemotherapy is increasingly used for early eradication of micrometastases to improve cure rate. The successful implementation of these two highly promising treatment strategies (organ preservation and total neoadjuvant therapy) hinges on the precise knowledge of (1) which patients will have a pathologic complete response; and (2) which patients will develop distant metastasis. Unfortunately, current clinical tools are rather crude and do not allow for accurate outcome prediction on an individual basis, leading to over-treatment in some patients and under-treatment in others. There is a pressing unmet need for reliable prognostic and predictive biomarkers to guide personalized treatment of rectal cancer. To address this unmet need, we adopt a rational approach for prognosis prediction by developing radiomic and deep learning models that are informed and guided by established knowledge of the tumor pathobiology. Additionally, we propose novel approaches to analyze serial images for predicting pathologic response to neoadjuvant therapy. Further, by leveraging the complementary value of imaging and blood-based biomarkers, we will construct integrative models to improve outcome prediction. To establish clinical validity, we will use a large retrospective dataset for model training and conduct rigorous prospective validation. If successful, this project will lead to risk-adaptive and response-driven personalized treatment strategies, which may ultimately improve outcomes for patients with rectal cancer.

NIH Research Projects · FY 2026 · 2024-01

SUMMARY Human heart development represents an ideal model system for understanding 1) how normal development can produce all the cell types necessary for cardiac function and 2) how genetic variation can perturb this process and lead to disease. We will generate large-scale single cell data sets that will enable the development of accurate computational models capable of predicting the effects of both genetic changes to regulatory elements (REs) and perturbations to trans-acting regulatory factors on gene expression during the complex developmental process of human heart development. We will study a medically relevant, human, in vitro, temporally dynamic, 3D cardiac organoid differentiation system that faithfully recapitulates fetal differentiation patterns for differentiation towards various cell types, including cardiomyocyte, neural crest, and cardiac endothelium. For each of these differentiation trajectories, we will work in distinct aims toward mapping, perturbing, modeling, and model validation: Mapping: we will generate systematic, single cell multi-omic (RNA-seq, ATAC-seq, and protein quantification) and bulk data to map REs, chromatin contacts, RNA polymerase, and gene expression through differentiation of human induced pluripotent stem cells to heart tissue. Perturbing: We will use CRISPR- based methods to comprehensively perturb transcription factors (TFs) required for the different differentiation trajectories, and map the single-cell gene regulatory and expression impact of perturbing these factors at multiple time points across these differentiation. Modeling: We will develop multi-input, nucleotide-resolved neural networks to learn dynamic gene regulatory networks using these mapping and perturbation data sets. These models will aim to understand the changing landscape of regulation and grammars of TF motifs over differentiation time and will predict both chromatin and gene expression effects expected from genetic perturbations. Model validation: We will apply our network models to identify, investigate, and experimentally test perturbations relevant to understanding disease variation, by knocking down TFs, perturbing REs, and editing disease-associated noncoding variants. Finally, we will extract and test molecular properties of TF function from validated models. Successful completion of our project will provide mechanistic interpretations for how genetic variants may impact development (by disrupting REs that in turn disrupt gene expression) in human heart development. Our Stanford team comprises a collection of investigators with a history of collaboration and work in consortia, and with expertise in genomics methods development (Greenleaf, Engreitz, Bassik), single cell methods and analysis (Greenleaf), heart development and disease genetics (Gifford, Quertermous), and deep learning for genomic data sets (Kundaje). This project will produce gold-standard data defining the trans-acting factor network driving heart development, and a model capturing these complex dynamics capable of quantitatively linking changes in genotype to effects phenotype relevant to human disease.

NIH Research Projects · FY 2025 · 2024-01

PROJECT SUMMARY/ABSTRACT The central nervous system contains an enormous number of neurons that are interconnected to form the circuits governing all brain functions. Neural circuits are built following a series of developmental steps such as neurite guidance, target selection, and synapse formation. Cell surface proteins (CSPs) enable neurites to connect with the correct synaptic partners. CSPs are not static within the membrane. They undergo dynamics and can be cycled on and off of the membrane which influences how they respond to and control the physiological and signaling events underlying neurodevelopment. Endocytosis internalizes CSPs from the membrane into the cytosol and has dramatic effects on CSP signaling and function. However, CSP endocytosis has largely been investigated in the context of a few known dynamic CSPs and mostly in culture systems. Though these approaches are informative, we still do not know the identity and number of endocytosed CSPs within a given developmental context or how their dynamics impact circuit assembly in vivo. The Drosophila olfactory system displays stereotyped wiring and high genetic tractability, making it a powerful in vivo system to elucidate the role and regulators of dynamic CSPs in circuit assembly. Here, ~57 distinct types of olfactory receptor neurons (ORNs) project their axons from the periphery into the brain to make one-to-one synaptic connections with the dendrites of ~57 corresponding types of projection neurons (PNs). This one-ORN-type-to-one-PN-type motif creates ~57 distinct and invariant information relay channels called glomeruli. In fact, impairing CSP dynamics disrupts olfactory circuit assembly, as well as the development of other sensory systems. The proposal will provide foundational insight into how CSP dynamics govern olfactory circuit assembly and enable the candidate to build an independent research direction to pursue in her own lab. Specifically, the K99 phase of this application focuses on: 1) employing cutting-edge quantitative proteomics to profile dynamic CSPs in ORNs and PNs in the developing brain; and 2) performing functional analyses to determine how CSP endocytosis impacts circuit formation. This work will be completed under the guidance of world-renowned neuroscientist, Dr. Liqun Luo (mentor), expert cell biologist, Dr. Kang Shen (co-mentor) and in collaboration with leaders in proximity labeling, Dr. Alice Ting, and mass-spectrometry, Dr. Steven Carr. The R00 phase of this proposal seeks to evaluate the intracellular mechanisms that control CSP dynamics by building upon expertise gained from the mentored phase of this proposal. The research in the mentored and independent portions of this application is directly in line with NIDCD’s mission to understand chemo-sensation and treat chemosensory disorders because elucidating the cellular mechanisms controlling olfactory circuit assembly could enable us to ameliorate sensory circuit defects that arise from disorders, injury, disease, or ageing.

NIH Research Projects · FY 2026 · 2024-01

Project Summary Despite concerted efforts to develop an effective vaccine over the last 40 years, the HIV pandemic remains a profound global public health challenge, with 75 million people infected and an estimated 32 million deaths from AIDS-related illnesses since 19811. The discovery of patient-derived broadly-neutralizing antibodies (bnAbs) has raised hopes that the immune system is capable of producing antibodies that might prevent HIV-1 infection2. However, efforts to elicit similar antibodies through vaccination schemes have been stymied, likely due to the combination of requirements for rare precursors, extensive somatic hypermutation (SHM), and uncommon heavy-chain complementarity determining region 3 (CDRH3) lengths in previously targeted bnAbs2. Recently, a novel epitope centered on the heavily glycosylated “silent face” (SF) of gp120 was found to be targeted by multiple antibodies derived from different germlines that were capable of cross-clade neutralization of tier 1, 2, and 3 HIV-1 viruses3–5. Importantly, the two glycans that form a majority of the SF epitope are >85% conserved, suggesting that antibodies targeting this epitope can in theory achieve neutralization breadths comparable to bNAbs against the highly-conserved CD4 binding site. Importantly, the relatively limited SHM (~20% amino acid) and ability to arise from multiple germlines in different donors suggests that SF-targeting antibodies may represent a more achievable target for elicitation via germline-targeting vaccination approaches; additionally, these antibodies also possess a complementary breadth and potency to the more well studied V3-targeting bnAbs making them an attractive addition to a potential poly-epitope targeted vaccine immunogen. Thus, the central goal of this proposal is to develop HIV-1 Envelope (Env) trimer immunogens capable of binding to the inferred germline versions of SF-targeting antibodies in vitro, which we hypothesize will be capable of eliciting a primary B cell response to the SF epitope in vivo. To test this hypothesis, I will (i) use yeast display to select for gp120 mutations that improve binding to SF-targeting germlines, (ii) express these mutations in stabilized trimeric Env immunogens to decorate nanoparticle vectors, and (iii) vaccinate knock-in mice transgenic for SF inferred germlines and characterize the naïve human B cell response to our germline-targeting immunogens. The long-term goal of this work is to validate the SF epitope as a target for germline-targeting vaccine approaches. Further outcomes include a better understanding of the germlines capable of responding to this epitope, which could help guide further rational immunogen design for an HIV vaccine.

NIH Research Projects · FY 2026 · 2024-01

PROJECT SUMMARY AND ABSTRACT: Cellular innate immune receptors play a vital role in recognizing pathogenic molecules and initiating an innate immune response to combat infections. However, when these receptors mistakenly sense "self" molecules, it can lead to the development of autoimmune diseases. One such cytosolic receptor, MDA5, forms filaments on viral double-stranded RNA (dsRNA) to activate the antiviral interferon (IFN) signaling pathway. The question arises: how does MDA5 differentiate between "self" cellular dsRNAs and "non-self" viral RNAs? Research con- ducted by our team and others suggests that the ADAR1 protein catalyzes Adenosine-to-Inosine (A-to-I) RNA editing on cellular dsRNAs, marking them as "self" RNAs and preventing erroneous MDA5 sensing. Mouse mod- els deficient in ADAR1 RNA editing are not viable, but their survival to the full lifespan can be rescued by remov- ing MDA5. In humans, rare autoimmune diseases such as Aicardi-Goutieres Syndrome exhibit loss-of-function mutations in ADAR1 and gain-of-function mutations in MDA5. Our recent work has shown that insufficient editing of dsRNAs is associated with presumed dsRNA-mediated inflammation in various autoimmune and inflammatory diseases, including Inflammatory Bowel Disease, Multiple Sclerosis, Parkinson's Disease, and Coronary Artery Disease. In the well-established ADAR1-dsRNA-MDA5 axis, there is an urgent need to identify the specific cel- lular dsRNAs that require ADAR1 editing to evade MDA5 sensing. We define these dsRNAs as immunogenic dsRNAs, as their lack of editing leads to increased immunogenicity. The primary objective of this research is to develop effective approaches to systematically identify and charac- terize immunogenic dsRNAs involved in inflammatory diseases and to investigate how the editing status and expression level of dsRNAs influence their immunogenicity. Firstly, we will utilize a recently developed genetic approach to identify immunogenic dsRNAs in various human cell types that we determine to be most relevant in inflammatory diseases. Subsequently, we will employ genetic, biochemical, and computational methods to thor- oughly characterize and validate these immunogenic dsRNAs. Secondly, we will expand our work on systemat- ically identifying genetic variants associated with editing levels. This expansion will enable us to pinpoint immu- nogenic dsRNAs that colocalize with GWAS loci associated with common autoimmune and inflammatory dis- eases. Additionally, we will develop a predictive model to assess an individual's burden of immunogenic dsRNAs as a cumulative measure. This model will aid in stratifying patients with inflammatory diseases, where impaired dsRNA editing and heightened dsRNA sensing likely contribute to chronic inflammation. In summary, by identi- fying the most crucial dsRNA substrates of ADAR1 and ligands of MDA5, this work will bridge a significant knowledge gap in the field. Furthermore, it will provide important mechanistic insights and therapeutic potential regarding how ADAR1 RNA editing prevents MDA5-mediated dsRNA sensing and innate immunity in both healthy and inflammatory disease contexts.

NIH Research Projects · FY 2025 · 2024-01

Project Summary Type 1 diabetes (T1D) is a chronic disease with a demanding self-management burden. Few (<25%) adults with T1D meet ADA-recommended targets for glycemic control, thereby increasing long-term complication risk. Diabetes technology (insulin pumps; continuous glucose monitoring, or CGM) alleviates management burden and promotes achievement of treatment goals. CGM systems improve glycemic control, benefitting long-term health outcomes without increasing risk of hypoglycemia. Major improvement in CGM accuracy has led to advances in closed loop systems that integrate an insulin pump and CGM to partially automate insulin delivery, improve glycemic control and reduce T1D management burden. To reap these health benefits, CGM users must wear the device daily. Despite benefits, a concerning proportion of adults who try CGM later discontinue use, but remaining drivers of discontinuation and inconsistent use are unclear. Greater understanding of when and why adults with T1D discontinue CGM after initial adoption is needed to target and tailor well-timed interventions to promote and sustain consistent CGM use. The proposed study complements the parent K23 (DK119470) goal of refining and evaluating a behavioral intervention, ONBOARD, that aims to promote CGM adoption and sustained use through targeting top known modifiable barriers (physical, data, social, and trust) for adults with T1D. Preliminary data from the parent trial show an estimated 24% of CGM adopters discontinuing use. This mixed method study extends the parent K23 by gathering in-depth qualitative data on participants reporting CGM non-use 12-months after initial CGM adoption; and conducting longitudinal analysis on factors linked to reduced CGM use and discontinuation. ONBOARD participants receive 3 months of CGM supplies to enable initial adoption. We collect CGM, A1c, and psychosocial data over 12 months, creating opportunity for systematic investigation of CGM use and discontinuation following uptake. The specific aims in this proposal are to 1) collect in-depth qualitative data on key remaining drivers of CGM discontinuation through user-centered methods, and 2) identify factors associated with lower CGM use and discontinuation in the year following CGM initiation using the full K23 sample. Examining remaining modifiable barriers to durable CGM uptake in adults with T1D will expand the reach of the existing K23 study to optimize timing and content of support for individuals at risk of discontinuing CGM. Completion of the proposed R03 aims will generate critical preliminary data to conduct a large scale R01 study to test a refined ONBOARD intervention to promote uptake and prevent discontinuation of CGM in adults with T1D. Completion of these proposed aims will further Dr. Tanenbaum’s career goal to establish an independent programmatic line of research focused on optimizing diabetes technology use to improve health and quality of life outcomes for adults with T1D.

NIH Research Projects · FY 2026 · 2024-01

Visual sensory substitution devices, in which images from a video camera are converted to a cross-modal signal that is then presented to the subjects in place of direct visual information, offer an alternative, non-surgical approach for the blind to appreciate aspects of their immediate environment. These non-invasive, low-cost technologies can also offer higher visual acuity and wider fields of view than currently available retinal and cortical implants. However, sensory substitution devices are rarely adopted to interpret and translate complex, natural environments for daily use by blind individuals partly because of their impracticality. In addition, little is known about the perceptual and behavioral consequences of delivering new patterns of information as an alternative sense when the visual system is damaged at different ages. The metabolic and functional processes that allow the deprived human visual cortex for enhancing cross-modal plasticity have not been clearly defined either. These research gaps must be filled before sensory substitution can be exploited as a method of vision rehabilitation for improving function and independence while reducing the associated costs of blindness. The goal of this project is to develop and refine sensory substitution technologies and to identify the determinants of cross-modal plasticity in brains deprived of visual input in order to facilitate sensory substitution. To achieve this goal, we will incorporate artificial intelligence (AI) to simplify images of complex, natural environments into isolated objects for sensory substitution. We will then use magnetic resonance imaging (MRI) and spectroscopy (MRS), combined with behavioral assessments to examine the neural substrates of sensory substitution in early and late blind subjects. We will leverage tactile and auditory substitution stimuli to determine how changes in neurochemicals reflect the cross-modal activity of visually deprived brains. We will also test how training with AI can help blind individuals interpret visual environmental cues in a more meaningful way. Feedback from the behavioral and neurobiological results will help refine our sensory substitution devices. Aim 1: To improve sensory substitution toward practical use, we will use AI and 3D coding to convert complex everyday scenarios into simplified and aesthetically engaging versions that aid both immediate task performance and sensory substitution training. Aim 2: To determine how visual processing pathways in the brain adapt to vision loss, we will employ advanced MRI and MRS to unveil the structural, metabolic, and functional brain properties in participants with different onset ages and durations of blindness before training. Aim 3: To determine the effects of sensory substitution training on visual processing pathways and performance of the blind, we will use advanced MRI and MRS to determine if cross-modal perceptual learning alters the deprived visual cortex toward a more plastic state via modulating its excitatory-inhibitory balance and choline levels. We will also examine the neurobehavioral changes that occur when identifying and locating the objects of interest using auditory or tactile stimuli with and without AI-assisted image deconstruction.

NIH Research Projects · FY 2025 · 2024-01

Project Summary/Abstract Mendelian variation in pigmentation is a rich genetic resource to identify and study signaling pathways relevant to human biology, but the repertoire of mutations that affect pigmentation in mice and in humans coat color has become saturated. Advances in genome sequencing technology now extend the reach of forward genetics beyond humans and model organisms to domestic and wild animals, for which there are many pigmentary phenotypes likely to represent conserved signaling pathways but whose molecular basis is unknown. Using pigmentary variation in domestic cats, our research group discovered a striking connection between a novel transmembrane protease encoded by Transmembrane aminopeptidase Q (Taqpep) and a secreted Wnt inhibitor encoded by Dickkopf 4 (Dkk4). Initial studies of Taqpep knockout mice reveal a role in multiple cell types and tissues including hair, the enteric nervous system, and the palate. In a different non-model organism, the plains zebra, Taqpep is a candidate gene for aberrant striping in an isolated population in Rwanda. Our results point to a central role for Taqpep in pathways and tissues that are directly relevant to human biology and disease, but a molecular understanding of those pathways is most effectively approached by combining forward genetics for gene discovery in non-model mammalian organisms with experimental genetics in laboratory mice and studies in cultured mammalian cells. We propose to: (1) Identify additional components of Taqpep action by forward genetics in Bengal cats and by analysis of fetal cat skin (2) Expand initial genetic studies of color pattern variation in zebras (3) Determine the mechanism of Taqpep action using experimental mouse genetics and proteomics

NIH Research Projects · FY 2025 · 2024-01

Project Summary. Cancer is governed by evolutionary principles whereby sequential changes at the genetic and epigenetic level enable proliferation, immune evasion, drug resistance, and metastasis. An outstanding goal in cancer biology is to understand the spatiotemporal processes underpinning this evolution. To do so would greatly improve our ability to create more effective treatment strategies and forecast tumor development far into the future. However, this goal remains elusive due to our incomplete catalog of molecular processes driving evolution and lack of molecular and computational tools for holistically profiling tumors. One emerging driver is extrachromosomal DNA (ecDNA). Found in approximately half of cancers and strongly associated with poor survival, ecDNAs are a unique form of oncogene amplification: they reside outside of chromosomes and exhibit elevated copy-number, gene expression, and chromatin accessibility as compared to chromosomal amplifications. New evidence has underscored the importance of ecDNA dynamics and heterogeneity in driving tumor progression: first, ecDNAs asymmetrically segregate during mitosis, leading to accelerated copy-number gains and rapid adaptation to stressful conditions. Second, several varieties of ecDNAs can exist in single cells where they form cooperative, intermolecular “hubs”. Despite this appreciation, these features of ecDNA have remained elusive to study due to a scarcity of tools for profiling their vast heterogeneity and stochastic evolutionary dynamics. In this project, I will develop the requisite computational tools for profiling ecDNA variability and evolution in cancer and use these tools to more thoroughly investigate how ecDNA heterogeneity is created, maintained, and leveraged in response to targeted therapies. First, I will build on breakthroughs in long-read sequencing to develop tools that enable unbiased multi-omic profiling of ecDNA variability across biological conditions (Aim 1). Second, I will leverage new molecular techniques to infer the phylodynamic properties of ecDNA lineages and learn the molecular fitness landscape of ecDNA (Aim 2). Third, I will explore the co-evolutionary principles of ecDNA by combining evolutionary modeling and CRISPR- based screens (Aim 3). Together, these studies will illuminate properties of ecDNA evolution, nominate new therapeutic strategies, and provide innovative computational tools for the greater scientific community. This work will be performed in the excellent training environment of Stanford University under the mentorship of Dr. Howard Chang, an expert in epigenomics and ecDNA. An advisory committee of leaders in the fields of ecDNA, cancer biology, bioinformatics, and tumor evolution will provide additional expertise and mentorship. The first half of each aim will be completed predominantly during the K99 phase of the award, providing a solid foundation for the aims in the R00 phase and eventually an independent R01 application.

NIH Research Projects · FY 2026 · 2024-01

Corneal blindness is a leading cause of visual impairment, affecting an estimated 12.5 million people worldwide. Corneal transplantation, or keratoplasty, is the only curative treatment for corneal blindness, yet the cadaveric donor tissue required for this sight-restoring procedure is available to less than 2% of patients worldwide. While several different tissue-engineered corneal transplants have been proposed, designing therapies with the long- term transparency and regenerative capacity of a human donor cornea remains a formidable challenge. The native cornea is a multi-layered tissue, with each layer playing a distinct role in the overall corneal function; however, most tissue-engineered corneas to date have been monolithic structures. Here we propose leveraging recent advances in 3D bioprinting to fabricate a bilayered, tissue-engineered cornea, with each layer fabricated from a customized biomaterial formulation with optimized print parameters to achieve the desired biofunctionality. This technology builds on our recently reported family of UNION bioinks that form cohesive, strong interfaces between distinct biomaterials when 3D-printed into a tissue-engineered construct. In Aim 1, we design a 3D-printed stromal layer with encapsulated corneal stromal stem cells (CSSCs) that resists contraction, remains transparent, and secretes pro-regenerative paracrine signals. While CSSCs have the potential to reduce scarring, their contractile phenotype leads to deformation of engineered tissue, altering the shape, size, and transparency over time. We hypothesize that covalent crosslinking of a chemically modified collagen type I bioink will enable printing of a mechanically robust hydrogel with ordered fibrils that guide cell alignment and phenotype. Our preliminary data demonstrate that bio-orthogonal azide-alkyne click chemistry crosslinking resists cell-induced deformation and promotes cell alignment. In parallel, in Aim 2, we design an acellular 3D-printed basement membrane layer that promotes corneal epithelial cell (CEC) migration and forms a cohesive interface with the 3D-printed stromal layer. CECs are known to alter their migration in response to properties of their underlying basement membrane, which is distinct from the stromal layer. Clinical translation requires a cohesive interface between these two layers to maintain a stable graft. We hypothesize that a thin layer of printed protein can be formulated to promote migration of endogenous CECs while maintaining a cohesive interface with the underlying stromal layer through bio-orthogonal covalent crosslinks. The bilayer construct will be evaluated in an ex vivo rabbit eye model to quantify the migration of endogenous CECs in response to secreted signals from the encapsulated CSSCs. In Aim 3, we evaluate the translational potential of the 3D-printed, bilayered corneal tissue in a preclinical rabbit model of keratoplasty. The rabbits undergo a 5.0- mm keratectomy to excise a central corneal stromal scar and will be treated with the full bilayer construct, the bilayer construct without CSSCs, and a monolayer construct without the engineered basement membrane. Injured, untreated animals are negative controls, and the uninjured eye in each animal are positive controls.

NIH Research Projects · FY 2026 · 2024-01

PROJECT SUMMARY Congenital heart disease (CHD) affects about 40,000 births per year (1%) in the United States. Estimates indicate that over 1 million children and 1.4 million adults with CHD are living in the United States. This project specifically aims to develop and clinically evaluate a magnetic resonance imaging (MRI) method to assess car- diovascular hemodynamics in pediatric CHD patients, when imaging is most needed. Individuals born with CHD require a lifetime of clinical care, especially during surgical planning and repair. This includes monitoring and/or serial interventions that mandate multiple echocardiography, CT, or MRI exams. Under most circumstances, MRI is the clinically preferred imaging modality for these patients owing to soft tissue contrast, clear depiction of complex anatomies, capability to quantify cardiac function, and the lack of ionizing radiation. Moreover, the ability to acquire 3D+time flow encoded (“4D-Flow”) MRI exams that depict and measure blood flow dynamics is an increasingly impactful clinical tool in adult patients, but several challenges persist in pediatric patients. Current pediatric cardiovascular 4D-Flow exams are slow to acquire (⇡20-min with moderate spatial and tem- poral resolution), have unacceptable flow measurement accuracy and precision (oftentimes >20%), can have long reconstruction times (several hours), and have inadequate clinical robustness (>10% of exams with unac- ceptable image quality). These are unacceptable clinical limitations that interfere with clinical care. To overcome these limitations, we have planned three specific aims to develop our Hyper 4D-Flow MRI exam. [AIM-1] To enable fast, accurate, and robust image acquisitions with Hyper 4D-Flow. [AIM-2] To enable fast and accurate image reconstructions of Hyper 4D-Flow. [AIM-3] To demonstrate the diagnostic value of “push-button” Hyper 4D-Flow in pediatric patients with CHD. Successful completion of the proposed specific aims will make available a validated, “push-button”, multi- vendor deployable Hyper 4D-Flow MRI exam that enables fast (5-min scan time and 30-min reconstruction), accurate (<5% bias), precise (95%-CIs<5%), and robust measurements (>95% of exams with diagnostic image quality; no need for post-acquisition corrections of background phase-induced errors) of blood flow. This compre- hensive approach to the problem will improve diagnostic accuracy and clinical confidence in decision making for pediatric patients with CHD patients, by providing fast, accurate, and robust evaluation of hemodynamics.

NIH Research Projects · FY 2025 · 2024-01

Project Abstract The nucleus accumbens (NAc) is a target for non-invasive brain stimulation for the treatment of substance use disorders (SUDs). Due to its location deep in the brain, transcranial ultrasound stimulation (TUS) is unique among non-invasive brain stimulation methods to be able to focus on the NAc with high spatial resolution. TUS has been shown to be effective in animal models, without any evidence of damage from either behavioral studies, H&E stained sections, or TUNEL staining. Further, there have been millions of diagnostic imaging studies at the pressures proposed for TUS, that have appeared to be safe. As a result, it is currently under active investigation in humans. However, when it comes to safety, we are compelled to examine all existing data. In particular, we have existing data from our prior studies in which we applied TUS to the sheep brain and examined it on H&E stained sections, as well as TUNEL staining. Here in this proposal, we seek to extend this analysis to the examination of inflammatory cytokines and the examination of glial cells using immunohistochemistry. These studies will give us a more complete picture whether TUS has an inflammatory effect on the brain.

NIH Research Projects · FY 2026 · 2023-12

Modified Project Summary/Abstract Section We propose that artificial intelligence (AI) screening for acute coronary syndrome (ACS) can improve ST-elevation myocardial infarction (STEMI) identification, and eliminate structured biases that disproportionately cause diagnostic delay in women and non-white patients. Every year 26 million patients visit EDs for ACS evaluation. All need an ECG in a critical 10-minute window of time after arrival to diagnose the most time-sensitive and life-threatening form of ACS, STEMI. These early ECGs occur before physician evaluation. In our prior work, every 1-minute reduction in time to ECG yielded a 1.24-minute reduction in the time to STEMI treatment, and every minute improves ACS clinical outcomes. Yet 37% of STEMI patients do not receive an ECG within 10- minutes. Those experiencing ECG delay disproportionately include women and non-white patients. These patients are twice as likely to die within 1 week compared to those receiving an ECG within the 10-minute window (11% vs 5%, p=0.02). These unacceptable findings need a diagnostic improvement solution. We will address this challenge by developing ACS screening artificial intelligence (AI) that will support ED care to 1) prevent missing STEMI patients by improving on the existing oversimplified manual screening practice, and 2) eliminate structured biases disproportionately impacting women and non-white patients for whom we will improve diagnostic precision. Our proposal aims to reduce disparities in timely care and mortality to improve access via 2 innovative contributions. First, we will test the desktop-to-bedside translation of a predictive model that out-performs manual screening. Second, we will improve the model by including considerations for risk variance and leveraging the calculation strength of machine learning to improve ACS prediction. Our goal is to reduce mortality for all by bridging a translational research gap. The impact of overcoming these practice limitations is that we will validate a pathway to reduce delays in STEMI identification. We will also close a clinical outcome disparity between women and men, and patients of non-white and white races. Our multidisciplinary team of experts in emergency medicine, interventional cardiology, predictive modeling, biostatistics, and clinical informatics have the expertise to successfully execute this proposal.

NIH Research Projects · FY 2026 · 2023-12

Modified Abstract Section Abstract PROJECT SUMMARY Statins are the cornerstone for preventing and treating atherosclerotic cardiovascular disease (ASCVD), the leading cause of death worldwide. Yet even the highest-risk patients face challenges initiating and continuing these low-cost, life-saving medications. Clinicians also fail to prescribe statins when indicated, and health systems may lack the infrastructure to systematically monitor statin adherence. Thus, the objective of the proposed study, Adherence Determinants in the Health Electronic Record Evaluation of Statins (ADHERES), is to determine reasons for statin nonuse as a strategy to close treatment gaps and improve clinical outcomes. Specifically, this study will fill the knowledge gap around patient, clinician, and system level factors impacting statin adherence by using electronic health record (EHR) data and focus groups. Statin use is typically collected in discrete, structured EHR data fields. However, structured EHR data may be incomplete, lack vital information about shared decisionmaking available around statin use, and fail to capture the variety of reasons for statin nonuse. A better understanding of the patient, clinician, and health system factors associated with statin nonuse thus requires the simultaneous study of structured, unstructured, and patientgenerated EHR data from different data sources, and ascertaining patient and clinician perspectives through qualitative approaches. To tackle this pressing health priority, a multidisciplinary team that includes experts in preventive cardiology, biomedical informatics, implementation science, and qualitative methods has been assembled. EHR data sources include ~3 million eligible patients from Stanford Health Care, the Veterans Health Administration, and Houston Methodist. Aim 1 will characterize gaps in guideline-directed statin use across patient and clinical indications using structured and unstructured EHR data and identify patient profiles and characteristics linked to statin nonuse. Aim 2 will elucidate reasons for statin nonuse in unstructured EHR and patient-generated health data using frontier large language models (LLMs). Aim 3 will validate and extend these reasons and co-design solutions to improve statin adherence through 10 patient (n=80 participants) and 4 clinician (n=20 participants) focus groups. Findings from ADHERES will be directly actionable in all three health systems by more accurately capturing gaps in guideline-directed statin use, and lead to the implementation of user-centered interventions to address these gaps. This multipronged approach can be scaled beyond statins to improve adherence and the use of other guideline-directed, lifesaving cardiovascular treatments.

NIH Research Projects · FY 2026 · 2023-12

PROJECT SUMMARY Regulatory DNA elements, including enhancers and promoters, encode multiple transcription factor binding sites (TFBS) that quantitatively tune gene expression in a cell-type specific fashion. Understanding and engineering regulatory DNA could unlock new therapeutic approaches — for example, to restore proper expression of a disease gene. Diseases of haploinsufficiency, such as Alagille Syndrome, are one example where such an approach could be transformative. Alagille Syndrome involves haploinsufficiency of JAG1, where improper dosage in vascular endothelial cells and smooth muscle cells leads to life-threatening complications including biliary atresia as well as right-sided congenital heart defects. An ideal gene therapy solution for Alagille Syndrome would be to engineer the promoter of JAG1 to turn up the production of the unaffected allele of the gene by 2-fold. Yet, our knowledge of how to program regulatory DNA to control gene expression is incomplete, in large part because we have lacked tools to accurately identify, edit, or characterize TFBS in regulatory elements in the genome. Our team has now developed innovative tools to dramatically increase the throughput of characterizing both endogenous and synthetic TFBS and their effects on gene expression in the context of Alagille Syndrome — allowing us to design DNA edits, test their impact on gene expression with high-throughput screens, and evaluate their ability to correct pathological patterns of gene expression. In Aim 1, we will build a genome-wide nucleotide-resolution map of TFBS in endothelial cells and smooth muscle cells from healthy and Alagille hiPSCs in conditions of static, physiologic, and pathological flow. In Aim 2, we will apply new pooled prime editing technologies to systematically mutate regulatory DNA sequences in the JAG1 promoter and identify TFBS that can increase JAG1 gene expression. In Aim 3, we will use CRISPR to correct JAG1 expression in hiPSCs from Alagille patients and characterize their effects on gene expression and cellular phenotypes. Together, these studies will illuminate basic mechanisms of Alagille Syndrome, test whether restoration of JAG1 function in cells from Alagille patients is sufficient to correct disease-associated cellular phenotypes, and demonstrate a new strategy to program gene expression in the human genome by combining nucleotide- resolution computational modeling with high-throughput sequence editing of endogenous gene promoters. This approach is generalizable and can be applied to other diseases where programming gene expression is desirable, such as for other haploinsufficiencies.

NIH Research Projects · FY 2026 · 2023-12

PROJECT SUMMARY Psychological studies have shown that social judgments from faces are ubiquitous, automatic, very rapid, and play critical roles in human social behavior, in both health and disease. The proposed project seeks to understand the neural mechanisms responsible. The goal of our study is to go beyond merely identifying the brain regions where social attributes of faces are represented, and to discover the mechanisms whereby we go from perceptual representations to social knowledge. To achieve this goal, we leverage the high temporal resolution of our recordings, the broad coverage of relevant brain areas that will be sampled, and the innovative analysis tools we have developed. The strengths of our data- driven approach will be as follows: (i) Predictive modeling with variance partitioning as a function of time will reveal what aspect of social judgments is attributable to visual features (shared variance). We will identify the specific set of visual features that are predictive and establish a mapping between these visual features and social representations as a function of time for each brain area, thereby providing fine-grained data on how social attribution representations emerge at specific points in time and in specific brain regions. This approach is designed to elucidate the mechanisms that bind perceptual face information with social knowledge. (ii) Contextual and electrical stimulation manipulations will experimentally perturb the relationship between visual features and social representations. Geometric analysis of the structure of the population code allows us to assess how specific social judgments are suppressed or amplified, thereby establishing a mechanism for how context changes the mapping between faces and the social attributions we established as part of (i). (iii) Decoding analysis will assess what aspect of the visual-social mapping is predictive of behavior in a given trial, thereby identifying a mapping between internal representations of social attributions and the actual choices patients make in a given trial. (iv) Inter-areal interaction analysis and latency comparisons will establish when information about a given social attribution is first available and how this information spreads to other anatomical nodes. Simultaneous recordings from multiple brain areas present a rare and unique opportunity to assess the role of directional interactions. We will use complementary measures of inter-areal spike-field coherence, Granger causality analysis, phase coupling, lagged correlation (during rest and experimental conditions), and dynamic resting-state iEEG connectivity as well as the measure of repeated single-pulse stimulations (also known as cortico-cortical evoked potentials). Our project will provide a new and dynamic picture of the brain where populations of individual cells in a distributed network interact together in time and space to enable the human brain to extract socially relevant information from faces

NIH Research Projects · FY 2025 · 2023-12

PROJECT SUMMARY/ABSTRACT Chronic hypertension in pregnancy is a growing contributor to adverse obstetrical and neonatal outcomes, but treatment with antihypertensives has been controversial due to unclear potential benefits and harms. However, in 2022, a large randomized controlled trial reported benefits of treating chronic hypertension in pregnancy to achieve blood pressure targets that are recommended for nonpregnant people (<140/90 mm Hg) without increasing risk to the fetus. Shortly thereafter, leading obstetrical societies in the U.S. issued recommendations to align with the trial findings. The long-term goal of the proposed project is to generate knowledge that improves the quality and equity of clinical care for pregnant and postpartum people with cardiovascular disorders and ultimately prevent morbidity and mortality. The overall objective is to leverage a nationwide distributed data network of electronic health records and administrative claims to inform optimal therapy for chronic hypertension in pregnancy. The central hypothesis is that type of medication, initiation timing, dose, and adherence affect outcomes of chronic hypertension treatment in pregnancy, which further differs with concomitant comorbidities. The rationale for this project is that recently revised national recommendations will likely increase the use of antihypertensive therapy for chronic hypertension in pregnancy, and the findings from this study will provide critical evidence on optimal therapy. The central hypothesis will be tested through three specific aims: 1) Compare the overall effectiveness and safety of the most commonly used medications to treat chronic hypertension in pregnancy; 2) Assess the impact of timing of initiation, dose, and adherence on the benefits and safety of treating chronic hypertension in pregnancy; and 3) Evaluate the benefits and safety of treating chronic hypertension in people with concomitant comorbidities, including multifetal gestation, severe obesity, preexisting diabetes mellitus, and chronic kidney disease. The proposed project is innovative by applying a target trial framework and advanced causal inference methods in a distributed data network created by the candidate. The research project is significant because optimizing treatment of chronic hypertension in pregnancy has the potential to benefit both the short- and long-term health of many pregnant people and their infants. This work will also be critically informative for future research led by the candidate on improving the treatment of preeclampsia, and hypertensive disorders during both the antepartum and postpartum periods. The proposed research and training will be accomplished with an exceptional mentorship team at Stanford University. The candidate will build on a strong background in perinatal epidemiology to gain expertise in perinatal pharmacoepidemiology, medical informatics and hypertensive disorders in pregnancy and postpartum. Together, the research and training will advance the candidate’s goal of leading an independent research program that improves the cardiovascular health of pregnant and postpartum people.