Stanford University

NIH Research Projects · FY 2025 · 2023-09

PROJECT SUMMARY While acute pain is an important biological signal in response to injured tissue, chronic pain occurs when the pain signaling outlasts the initial injury and has deleterious effects on health and quality of life. Chronic pain represents an enormous public health burden with few therapeutic options. Chronic pain is distinct from acute pain with several unique features including long-lasting activation of astrocytes. Astrocytes are CNS cells with diverse functions including energy homeostasis, regulation of the blood brain barrier, clearance of neurotransmitters, and regulation of synaptic transmission, all of which are altered in activated states. Preventing activation of astrocytes represents a key therapeutic target. Microglia, the resident immune cells of the CNS, have been implicated as key mediators of astrocyte activation. In this way, microglia manipulation may provide a tool to prevent or alter astrocyte activation, which in turn may prevent pain from becoming chronic. Previous research from our lab has shown that depletion of microglia at the time of transition from acute pain to chronic pain prevents chronic pain. However, when microglia are depleted once chronic pain is established, there are only transient improvements in pain-like behaviors. One explanation for these different effects is that microglia may be contributing indirectly to chronic pain by triggering astrocyte activation during the transition to chronic pain. However, once astrocytes are activated, microglia cease to have an active role in pain signaling and the alterations in spinal cord circuits maintaining chronic pain are due to changes in astrocyte function. I hypothesize that at the acute-to-chronic transition microglia are necessary and sufficient to activate astrocytes and that microglia effects in chronic pain are entirely dependent on astrocyte activation. In Aim 1, I will characterize astrocyte activation in a mouse model of pain-producing peripheral injury after selective depletion of microglia at the acute-to-chronic pain transition. I will further use DREADDs to exogenously activate astrocytes at the acute-to-chronic transition to determine if exogenous activation of astrocytes is sufficient to maintain the transition to chronic pain in the context of microglia depletion. In Aim 2, I will use exogenous activation of microglia in a naïve mouse to determine if activation of microglia is sufficient to activate astrocytes and if this activation of astrocytes leads to pain-like behaviors. Finally, in Aim 3 I will determine which signals from microglia are important for astrocyte activation in the induction of chronic pain using cell-cell interaction analyses of single nuclei RNA-Sequencing data from astrocytes and microglia. Using the innovative experiments in this research proposal, I will uncover the relative roles and contributions of microglia and astrocytes to chronic pain and generate new targets for pain therapeutics. The proposed research is part of a comprehensive training plan that will provide me with extensive technical knowledge in cutting-edge technologies and analyses. In addition, the career and scientific goals of this proposal will serve as the foundation for my future K99/R00 Pathway to Independence award and an academic research career.

NIH Research Projects · FY 2025 · 2023-09

PROJECT SUMMARY Many disease-associated variants in coding regions of the genome affect translated protein and enzyme products by perturbing their folded conformation or their function, such as interactions with substrates or macromolecular partners. However, we lack a unified predictive framework to predict functional effects of coding variants, limiting how genomic data can be used in precision medicine. Machine learning models trained on large sequence databases have claimed to predict deleterious effects from coding variants in several model proteins, but to date their practical usage has been limited because of two major challenges. The first is the lack of descriptive, “ground truth” biophysical datasets relating sequence variation to native protein properties, due to the low throughput of traditional biochemical and biophysical experiments. The second is that there is not a well- established method for integrating these data in state-of-the-art predictive models. To address these critical limitations, I propose to apply cutting-edge microfluidic techniques to generate large quantitative biophysical datasets connecting sequence variation to function in human acylphosphatase (ACYP), a model protein of the alpha/beta fold family (found in ~10% of human proteins), and leverage these data to enhance predictive models. This microfluidic platform (HT-MEK) contains an array of chambers that allow for parallel expression and purification of >1,700 proteins, and provides measurements of in vitro kinetic and thermodynamic constants for each. In Aim 1, I will engineer a series of ACYP functional assays using HT-MEK and derivative microfluidic technologies, first testing in vitro expression, on-chip stability, and catalytic turnover of a small library of ACYP variants and finally comparing to traditional biochemical measurements. In Aim 2, I will rapidly generate scanning mutagenesis libraries in ACYP and make measurements across hundreds of ACYP variants on HT-MEK. In Aim 3, in collaboration with ML experts, I will use this unprecedented quantitative biochemical dataset to fine-tune a cutting-edge deep learning to provide the first variant effects predictor enhanced by in vitro data at scale. My preliminary data has shown that this model can generate de novo ACYP sequences that fold and are catalytically proficient, suggesting that it will provide a strong foundation for functional prediction. Together, my results will provide insight into the utility of in vitro, biochemical datasets from human proteins in training better predictors of disease phenotypes. The training that I will obtain in carrying out these Aims will allow me to (1) develop skills in research design, analysis, and interpretation of protein biophysics data; (2) learn advanced techniques in protein biochemistry and statistical sequence analysis; and (3) obtain a competitive post-doctoral fellowship with the long-term goal of establishing an independently-funded laboratory at a research-intensive university.

NIH Research Projects · FY 2025 · 2023-09

PROJECT SUMMARY Genome-wide studies have now identified hundreds of thousands of associations between genes or genetic loci and human phenotypes, each of which could reveal mechanistic insights about disease biology. Yet, the cell-type specific functions of most of these genes remain unknown, and we currently lack the ability to connect these genes into cellular programs and thereby reveal the pathways important for disease. To address this limitation, our proposed Data Analysis and Validation Center aims to work together with the MorPhiC Consortium to build a Catalog of Cellular Programs — i.e. a map of which genes work together in biological pathways and their corresponding multimodal molecular and cellular phenotypes, in defined cell types or states. Our team brings a diverse set of expertise in computational genomics, methods and technology development, experimental design, interdisciplinary collaboration, consortium organization; and includes new junior investigators who will bring new forward-thinking ideas and tools to MorPhiC. We have developed a wave of innovative methods integrating CRISPR, single-cell, imaging, and human genetics data that will enable building such a Catalog of Cellular Programs and applying this Catalog to understand the genetics of human disease. The goals of our Center are to: (i) Define single-layer phenotypes, by applying a suite of computational state-of-the-art approaches for analysis and modeling of RNA, ATAC, and imaging data; (ii) define a multi-modal representation of molecular and cellular phenotypes, by identifying modules of features that co-vary across perturbations and single cells; (iii) build a Catalog of Cellular Programs that links genes to the molecular and cellular phenotypes they control, by inferring causal gene regulatory networks from perturbation data; (iv) apply the Catalog of Cellular Programs to demonstrate its utility identifying causal genes and programs for human diseases; and (v) participate in Collaborative Activities with MorPhiC, including to guide experimental design and ensure utility, robustness, and interoperability of Phase 1 datasets. Together, these aims will develop novel computational toolkits to infer causal gene regulatory networks from multi-modal perturbation data; construct a Catalog of Cellular Programs as a foundational resource for MorPhiC and the broader community; and demonstrate the utility of this Catalog through application to understand the genetics of human diseases.

NIH Research Projects · FY 2025 · 2023-09

Abstract Osteoarthritis (OA) is a major disease affecting 1 in 6 adults above 60 years of age in US that significantly impairs quality-of-life by impacting movement and function. Tissue health and disease is frequently governed by a complex and non-linear interplay of cell-intrinsic and systemic factors including both biochemical and biophysical cues. The overall aim of this project is to understand how the changes in ECM (extra cellular matrix) viscoelasticity affect cartilage homeostasis in health and during disease initiation and pathogenesis in OA. Recent studies by our team have elegantly demonstrated that ECM viscoelasticity governs cell volume in cartilage cells i.e. chondrocytes. Previous studies of cartilage biology had only examined the impact of ECM elasticity (i.e. “stiffness”), and the role of viscoelasticity had been mostly ignored. We found that viscoelastic hydrogels that exhibit fast stress relaxation, or were more viscous, could provide a microenvironment that is more conducive to anabolic gene expression in human chondrocytes resulting in increased ECM production, promoting a healthy chondrocyte phenotype. The underlying cause was observed to be the ability of chondrocytes to expand their volume in the fast relaxing gels, an ability that was restricted in the slow relaxing gels, which are more elastic. Understanding the optimal ECM viscoelasticity for healthy and human induced pluripotent stem cell derived chondrocytes can guide ideal scaffold preparation for cartilage tissue engineering. The aim of this proposal is therefore to optimize hydrogel viscoelasticity for engineering inflammation- suppressive cartilage constructs. We will firstly optimize development of cartilage constructs in fast relaxing hydrogels in the presence of dynamic mechanical loading. Secondly, these constructs will be tested in human and rat models of cartilage defects. Thirdly, we aim to gain an understanding of the molecular pathways underlying the relationship between mechano-transduction and inflammation in cartilage health and disease. The experimental outcomes from these studies have the potential to enhance therapeutic strategies for cartilage regeneration and OA that remain unmet clinical needs.

NIH Research Projects · FY 2025 · 2023-09

Cell proliferation is a fundamental biological process that often occurs for cells in a 3D context in vivo, in which cells are surrounded by extracellular matrix (ECM) and other cells, and various applications rely on the proliferation of cells within a biomaterial. It has long been known that changes in matrix stiffness impact cell behaviors through mechanotransduction, and mechanisms of stiffness-sensing in 2D culture are now established. However, the mechanisms mediating the impact of changes in matrix stiffness on cell proliferation in 3D remain unclear. Further, living tissues and ECMs are viscoelastic, exhibiting some characteristics of elastic solids and some of viscous liquids. Matrix viscoelasticity is sensed through mechanotransduction, and we have found that changes in matrix viscoelasticity impact cell spreading, migration, proliferation, stem cell differentiation, matrix deposition, morphogenesis, and gene expression. However, the mechanisms mediating the impact of matrix viscoelasticity on these processes, particularly proliferation remain unclear. The goal of the proposed work is to determine the mechanism mediating the impact of matrix stiffness and viscoelasticity on cell proliferation in 3D matrices. Our overall hypothesis is that mechanosensitive ion channel-mediated pathways and integrin-mediated pathways interplay to sense matrix viscoelasticity and stiffness, and subsequently control proliferation through changes in chromatin accessibility, YAP-independent transcription, and a set of molecular regulators not implicated from 2D culture studies. We will address this hypothesis in 3 aims, using an approach that involves the use of alginate hydrogels with independently tunable viscoelasticity, stiffness, and RGD ligand density for 3D culture of adherent cells, including fibroblasts, epithelial cells, and mesenchymal stem cells. In aim 1, we will determine the biophysical mechanisms underlying the impact of hydrogel viscoelasticity, stiffness, and adhesivity on the proliferation of adherent cells in 3D culture. In Aim 2, we will elucidate transcriptional and epigenetic regulation of mechanotransduction and proliferation, using RNA-seq and ATAC-seq combined with advanced bioinformatics analyses. In Aim 3, we will identify novel regulators of proliferation and mechanotransduction in 3D using genome-wide CRISPR screening. Innovative aspects of this approach include the study of mechanisms mediating mechanotrasduction and proliferation in 3D matrices, the focus on viscoelasticity (beyond stiffness), the potential for discovering YAP-independent mechanisms of mechanotransduction, the identification of how the epigenome regulates mechanotransduction and proliferation in 3D, and the application of a CRISPR screen to identify novel molecular regulators of mechanotransduction. The significance of this work is that it will determine the biophysical and molecular mechanisms by which ECM or biomaterial stiffness and viscoelasticity regulate cell proliferation in 3D. Given the importance of cell proliferation, the ubiquity of matrix viscoelasticity in ECMs, and the potential relevance of discovered mechanisms of mechanotransduction to other processes, the significance is expected to be high.

NIH Research Projects · FY 2024 · 2023-09

Project Summary Lack of transparency and trustworthiness of deep neural networks (DNNs) has long been recognized as a major drawback of the technology, hindering its widespread acceptance in many practical applications. The objective of this project is to establish a novel contrastive feature analysis (CFA) framework for reliable visualization of the high dimensional feature space and effective design of high-performance DNNs for medical image analysis. We hypothesize that CFA-based feature visualization will enable us to quantify the quality of the feature space at different layers during training/testing of a DNN and empower us with an effective tool to prune the network architecture for enhanced performance. Specifically, we will (1) develop an efficient visualization technique CFA for high dimensional feature data, 2) apply the CFA visualization framework to automatically refine DNN architecture for improved performance, and 3) demonstrate the potential of CFA in solving clinical problems. Successful completion of the project will enable us to analyze the feature data reliably and quantify the quality of the feature space at different layers of a DNN. The study also promises to provide high-performance DNNs for medical image analysis to substantially improve the AI-based diagnosis, prognosis and treatment planning of different diseases.

NIH Research Projects · FY 2026 · 2023-09

This application will combine the neuroimaging expertise of the investigator with new knowledge of biomarkers for aging pathology, advanced computational methods and Down syndrome clinical translational research. The goal is to produce a unique skill set which will be used to advance the candidate’s career as an independent lifespan developmental cognitive neuroscientist focusing on Down syndrome. Neurodegenerative changes associated with Alzheimer's disease (AD) are present in almost all individuals with Down syndrome by age 40 years, and the lifetime risk of developing dementia is more than 90% in this population. Understanding patterns of neuropsychiatric and pathological brain changes before AD diagnosis is critical in order to facilitate interventions prior to onset of irreversible neuropsychiatric and pathological brain changes. Yet charting of pathological brain changes and diagnosis of dementia is extremely challenging in individuals with DS who have brain and neuropsychological differences present through the lifespan. Here we propose a computational neuroscientific framework in which we will utilize conventional resting state functional MRI, advanced quantitative MRI and multimodal biomarker data to disentangle brain and neuropsychological changes specific to DS and AD (within the context of DS). Specifically, we will test the overarching hypothesis that whole brain connectivity (i.e. connectomes based on resting state functional MRI) and brain microstructure (quantitative MRI) represent intermediate phenotypes between primary brain pathologies and neuropsychological outcomes in DS and AD. First, we will leverage data from the Alzheimer's Biomarkers Consortium — Down Syndrome (ABC-DS) to identify connectome-wide signatures specific to DS, and a separate set of signatures specific to AD in the context of DS. Then, we will examine relationships between connectome-wide signatures and AD pathology within the amyloid, tau and neurodegeneration (ATN) framework. Next, we will examine relationships between connectome-wide signatures and critical neuropsychological functions including memory, social cognition, and executive function dimensionally. Finally, we will enroll a novel cohort of adults with DS and collect complementary advanced MRI studies (not available in ABC-DS) and fluid biomarkers, to examine brain microstructural properties which may be more a) sensitive to neurodegenerative changes when compared to traditional MRI and b) complementary to PET imaging. Together, the results of this proposal will advance a mechanistic understanding of DS- and AD-specific pathological brain changes and how reorganization of functional networks relate to neuropsychological changes. Further, these data form the basis for a follow-up R01 combining ABC-DS emerging longitudinal data with advanced complementary MRI and multimodal biomarkers.

NIH Research Projects · FY 2025 · 2023-09

PROJECT SUMMARY AND ABSTRACT Surgery places patients at increased for opioid use disorder and persistent postoperative opioid use (PPOU). In addition to its direct impact on patient health, PPOU, which affects 5-6% of surgical patients, is associated with an increased risk of opioid use disorder, opioid overdose, and surgical mortality/morbidity. This issue has particular salience for older adults. Over half of all surgical procedures in the United States occur among older adults, over half of older adults will require a surgery once in their lifetime, and the incidence of PPOU among older adults can be as high as 10%. In this light, the long-term goal of this project is to characterize the effectiveness of perioperative interventions in reducing the risk of long-term outcomes such as PPOU and opioid use disorder among older adults undergoing inpatient surgery. While a wide variety of interventions has been hypothesized to reduce the incidence of PPOU and other post-operative opioid outcomes, there remains a lack of consensus about their effectiveness, in part due to data limitations. In particular, it is often challenging to obtain detailed data on perioperative care (i.e., amount of opioid administered intraoperatively) and data on long-term opioid outcomes. This study builds on a novel dataset that links two datasets: the Multicenter Perioperative Outcomes Group (MPOG), a large, multicenter registry of surgical cases using data extracted from electronic medical records and a healthcare claims data for Medicare fee-for-service patients. This novel dataset unites the best aspects of both datasets: the ability to measure perioperative care and the ability to follow patients in order to assess long-term opioid outcomes. We will accomplish the goals of the project through four specific aims. First, we will augment the existing dataset by developing scalable and generalizable tools to incorporate relevant data from the inpatient stay (i.e. opioid administration and the use of non-opioid adjuncts). Second, we will demonstrate the feasibility of these methods at a single institution and expand their use to five institutions. Third, we will use the augmented dataset to evaluate the association between intraoperative interventions (i.e., opioid administration, use of nerve blocks) and long-term opioid outcomes (i.e., PPOU and opioid use disorder). Finally, we will use the augmented dataset to evaluate the association between inpatient stay interventions (i.e., reduced opioid utilization, reduced prescribing at discharge) and long-term opioid outcomes. The findings of this project will be significant as they will help guide crucial aspects of perioperative decision-making such as intraoperative and postoperative opioid administration. The expected outcomes of this project are relevant to the goals of the HEAL Initiative as they will enhance efforts to reduce the incidence of PPOU and other long-term opioid outcomes such as opioid use disorder. Crucially, throughout the project, we will work with stakeholders to maximize its impact, such as consulting with stakeholders to decide the interventions to study and working with stakeholders to incorporate the project findings into clinical guidelines and policy.

NIH Research Projects · FY 2026 · 2023-09

PROJECT SUMMARY/ABSTRACT Radiculopathy is a common spinal condition resulting from compression and irritation of the spinal nerve roots, leading to sensory deficits, muscle weakness, and pain. Dermatomal maps are a key component of the clinical exam and provide information on the correspondence between cutaneous sensations and the nervous system. Dermatomal sensory deficits can help localize neurological injury in spinal conditions and guide treatment. Dermatomal maps, however, are limited—they contain uncertainty in the neuroanatomy mapped (spinal nerve, dorsal root ganglion, dorsal horn, or spinal cord (SC) segment), assume left-right symmetry and no sex differences, provide no information on between-subject variability, and remain to be validated. SC functional magnetic resonance imaging (fMRI) permits the non-invasive in vivo spatial mapping of human SC activity. Here we will use SC fMRI to test hypotheses central to dermatomal maps, investigate the effects of neurological injury on SC sensory processing, develop markers of SC sensory activity, and test their diagnostic value in cervical radiculopathy while improving our SC fMRI methods. To accomplish this, we will first enhance our existing SC fMRI methods by building a research-grade 64-channel head-neck coil, testing a novel spatial normalization method that accounts for SC segment location, and exploring the use of surface electromyography to monitor and remove motor-related noise during fMRI experiments. We will compare the improved SC fMRI methods against our currently operational methods while characterizing the SC correlates of sensory stimulus intensity encoding using electrocutaneous sensory stimulation of the third digit of the right hand (C7 dermatome) in 30 healthy volunteers (HV) (20-79 years old, 15 females, 15 males). Then using the enhanced SC fMRI methods, we will quantitatively map the spatial distribution of SC activity in 120 right- handed HVs (20-79 years old, 60 females, 60 males, stratified by age) during electrocutaneous sensory stimulation of the first, third, and fifth digits (C6, C7, and C8 dermatomes, respectively) of the left and right hands. We will develop probabilistic maps of the spatial distribution of SC activity, assess the superior-inferior localization of activity, contrast the activity between left and right stimulation and sexes, and quantify between- subject variability. We will use machine learning algorithms to develop normative SC sensory markers by predicting the stimulation site. Finally, 40 right-handed patients with right-sided C7 cervical radiculopathy (30– 79 years old, 20 females, 20 males) and 40 age- and sex-matched HVs will also undergo the same SC fMRI experiment, and we will investigate group differences in SC activity to uncover the effects of neurological injury on SC sensory processing and then assess the diagnostic value of the SC sensory markers. Completing our aims will improve SC fMRI methods, validate/refute hypotheses central to dermatomal maps to better inform their use in clinical practice, advance our scientific knowledge of SC sensory processing in healthy and injured states, and provide preliminary validation of MRI-based diagnostic markers for cervical radiculopathy.

NIH Research Projects · FY 2025 · 2023-09

Hereditary hemorrhagic telangiectasia (HHT) is a genetic disease characterized by multiple arteriovenous malformations (AVMs) which are direct connections between arteries and veins, bypassing the capillary bed. Pulmonary AVMs (PAVMs) are the most common visceral AVMs in adult (10-45%) and pediatric HHT patients (60%) and cause significant morbidity and mortality due to an increased risk for cerebral abscesses, stroke, pulmonary hemorrhage and migraines. Current treatment for PAVMs consists of catheter mediated embolization with a re-perfusion rate of up to 25%, necessitating frequent imaging (radiation exposure) as well as repeat interventions. While heterozygous loss-of function mutations in ENDOGLIN, ALK1 and SMAD4 are responsible for the development of HHT in 85% of patients, we still do not know precisely how PAVMs develop. In particular, we do not know exactly from which vascular bed (arterial, capillary, venous) PAVMs arise, and which downstream signaling pathway is most important for PAVM development or growth that could be harnessed as a therapeutic target. No medical therapy exists that is able to prevent, arrest growth or even reverse PAVMs. Furthermore, we are lacking precise animal models of PAVMs or in vitro disease models, necessary for pre-clinical testing of therapeutic approaches. We therefore hypothesized that understanding the cellular and molecular mechanisms governing PAVM development paired with the identification of clinically relevant, pathological signaling abnormalities will allow us to develop and test novel therapeutic approaches that prevent and potentially reverse disease. Our proposal has three significant parts, which are represented by our three specific aims: First, to develop and characterize a novel mouse model of PAVM formation by deleting HHT causing genes in different endothelial cell subpopulations and study their role in PAVM development and growth. Second, to differentiate induced pluripotent stem cells (iPSCs) from HHT patients into arterial and venous endothelial cells (ECs), to identify novel common or unique pathways altered in HHT as a direct consequence of mutations in ENG, ALK1 and SMAD4, to predict repurposed drugs (in silico) and test whether they target the newly identified pathways in iPSC-ECs and tissue culture. Third, to test whether lead candidate drugs, FK506 and Enzastaurin, and novel drugs identified in Aim 2 (ie Brivanib, see preliminary data) positively influence PAVM formation, growth and potential regression. Our proposal is innovative because it combines a conceptionally novel approach (understanding PAVM development by focusing on disease-causing alterations in subpopulations of lung endothelial cells) with cutting edge techniques (multiplex single-molecule fluorescence in situ hybridization, spatial transcriptomics, multicolor labeling and high resolution 3-D imaging of the lung) and novel pharmacological interventions (drugs identified by High-Throughput Screening, predicting novel drugs in silico).The short-term impact will be a better understanding of how AVMs form in the lung and potentially in other organs (brain, skin). The long-term impact will be the identification of potential novel treatments for AVMs.

NIH Research Projects · FY 2026 · 2023-09

Digital technologies can have enormous impact in the prediction, early detection, and tracking of Alzheimer’s disease progression. In particular, there is a need to develop digital biomarkers that can detect early changes in brain function before the onset of cognitive symptoms and/or brain biomarkers. The EEG is a compelling candidate for an early “digital biomarker” of AD as numerous EEG features are known to be correlated with AD progression and fundamental biomarkers. Unfortunately, there is limited evidence that these same EEG measures, as currently constructed to describe population-level data, can accurately track, or predict AD progression in individuals. One reason for this is that EEG signals have many sources of with- and between- subject variation that are not accounted for in current analysis methods, leading to imprecise markers that only have sufficient statistical power at the population-level. There have been recent advances in neural signal processing that make it possible to account for these sources of error and in turn dramatically improve the precision of EEG-derived measures. Over the past several years our lab has made significant strides to account for these sources of error leading us to develop novel, sophisticated signal processing algorithms that can enhance the precision of EEG derived measures. Through the specific aims of this project, we seek to provide the AD research community with a suite of powerful, accessible signal processing software tools that will dramatically enhance the precision and quality of EEG-derived biomarkers related to AD progression.

NIH Research Projects · FY 2025 · 2023-08

PROJECT SUMMARY Obsessive-compulsive disorder (OCD) typically starts in childhood, leads to lifelong morbidity, and costs the economy $2.1 billion (direct costs) and $6.2 billion (indirect costs such as lost productivity) annually. OCD often responds inadequately to serotonin reuptake inhibitors (SRIs). Moreover, the typical SRI response is only partial and is delayed by 2-3 months. Given the dysfunction OCD imposes on millions of U.S. adults, exploring whether SRI non-responders are helped by adding drugs with different mechanisms of action is urgently needed. Emerging data support ketamine’s rapid and sustained anti-obsessional effects in OCD. We seek to understand the mechanisms underlying ketamine’s rapid OCD effect and thereby to speed development of more effective agents for OCD. For several reasons, this proposal focuses on ketamine’s action on the mu-opioid system and OCD neural circuits. First, our prior work found that the mu-opioid receptor antagonist naltrexone blocks ketamine’s acute, dramatic antidepressant effects. Second, naltrexone also worsens OCD symptoms, while morphine (a mu-opioid receptor agonist) diminishes OCD symptoms. Third, in major depression, attempts to develop follow- on agents that work via NMDA antagonism have not been successful. One possible explanation is that NMDA antagonism may not fully explain ketamine’s rapid and enduring antidepressant effects. To assess the neural targets modulated by mu-opioid receptor antagonism, we focus on fronto-striatal cognitive control circuits which neuroimaging studies show are directly influenced by the mu-opioid system. Indeed, mu-opioid receptors are distributed throughout the fronto-striatal circuits associated with OCD. Drawing on this rationale, the proposed mechanistic trial would be the first to probe the role of ketamine’s opioid properties in modulating fronto-striatal circuitry and bringing about anti-obsessional effects in OCD. We will test, by administering pre-treatment with naltrexone, whether opioid receptor antagonism blocks ketamine’s effects on control and reward circuits and circuit connectivity. Low-cost markers of opioid system engagement that predict anti-obsessional response are needed. We will therefore explore two promising, low- cost markers of opioid system engagement in humans. The proposed projects combine experimental medication and neuroimaging approaches in order to open a new avenue for therapeutics to transform psychiatric treatments.

NIH Research Projects · FY 2025 · 2023-08

PROJECT SUMMARY The largest cohort studies on the subject to date have found that adults with a low estimated sodium intake had a higher risk of mortality than those with a sodium intake close to the population’s average. These findings of a “J-shaped” sodium-mortality association have important implications for public health efforts on salt because they suggest that reducing sodium intake at the population level could cause harm among a large proportion of the population. Many investigators have criticized these cohort studies for their use of a spot urine, rather than the gold-standard 24-hour urine sample, to estimate sodium intake. Thus far, however, the debate on the shape of the sodium-mortality association could not be settled convincingly because analyses of cohort studies that collected 24-hour urine samples have suffered from far smaller sample sizes than of those that collected a spot urine. The overall objective of this project is to better understand the association of sodium intake with mortality, particularly at levels of sodium intake below the population average in the US. The central hypothesis is that the J-shape of the sodium-mortality association that has been reported by some investigators is a result of bias from assessing sodium intake through use of a spot rather than a 24-hour urine sample. This central hypothesis will be tested by pursuing three specific aims: 1) Determine the range of sodium intake among adults that is associated with the lowest risk of all-cause mortality; 2) Determine whether the commonly reported increased risk of mortality at low levels of sodium intake stems from using a spot urine (rather than the gold-standard 24-hour urine) to estimate 24-hour sodium intake; and 3) Establish whether the association of sodium intake with cardiovascular causes of death is the main determinant of the shape of the sodium-mortality association. Using a case-cohort design, this project will analyze the sodium and potassium concentration in urine samples collected as part of Lifelines, which is a population-based cohort study focused on healthy aging in the Northern Netherlands. The Lifelines cohort has three unique features that enable the investigators to make an important contribution to resolving the controversy about the J-shaped sodium- mortality association: 1) Lifelines has collected two 24-hour urine samples from its participants; 2) with 147,888 adult participants and over 6,000 deaths (over a 15-year follow-up period), the Lifelines cohort has a far larger sample size and number of deaths than any other cohorts that have collected 24-hour urine samples; and 3) Lifelines has additionally collected a spot urine sample from all participants such that we are able to directly compare the shape of the sodium-mortality association depending on whether a spot or the gold-standard 24- hour urine sample is used. This project is significant because it will inform dietary guidelines on salt, and whether public health efforts should attempt to lower sodium intake among the entire population (e.g., through reformulation of the food supply) or focus instead only on individuals who consume high levels of sodium.

NIH Research Projects · FY 2024 · 2023-08

PROJECT SUMMARY At the start of the COVID-19 pandemic in late 2019, an estimated 1.2 million people—including 158,500 (13%) with undiagnosed infection—were living with HIV in the United States. Since then, HIV control efforts have been complicated by disruptions to HIV testing, care-related services, and case surveillance activities in state and local jurisdictions. However, the full impact of the COVID-19 pandemic on HIV transmission, incidence and outcomes has been difficult to quantify. Wastewater-based epidemiology (WBE) is a non-invasive and unbiased surveillance approach that can be used to estimate infectious disease occurrence in the population by detecting pathogen DNA or RNA in pooled community samples of wastewater. Here we propose to apply a novel WBE HIV surveillance method to measure HIV-1 nucleic acids in wastewater to estimate HIV incidence in sewersheds during the COVID-19 pandemic. This research study will pursue three specific aims: (1) to develop and validate a quantification method for HIV-1 nucleic acids (RNA and DNA) in urine, feces and wastewater settled solids, (2) in 30 people living with HIV, to c orrelate HIV nucleic acid (RNA and DNA) shedding in urine and feces with plasma viral load, and (3) using archived samples of wastewater rom Santa Clara and San Francisco Counties during the COVID pandemic, to determine trends in wastewater HIV-1 nucleic acid levels and compare findings with community case rates of HIV. The overarching goal of this project is to establish an HIV quantification method for wastewater-based surveillance using digital droplet, reverse transcription-PCR analysis that can be used to monitor HIV in the community. We hypothesize that wastewater surveillance can identify populations disproportionately affected by HIV, facilitating allocation of resources to those at highest risk, thereby maximizing HIV control. Investigating rates of changes in HIV nucleic acid in wastewater in relation to COVID-19 may also improve our understanding of how pandemic disease and its control strategies can impact HIV surveillance and patient care. Knowledge gained from this project will help establish a framework for wastewater-based surveillance for HIV in the US and globally that can reduce health disparities, improve health outcomes and prevent HIV transmission.

NIH Research Projects · FY 2024 · 2023-08

PROJECT SUMMARY Learning and executing motor skills are crucial functions of the brain and involve the coordinated activity of multiple brain regions. Traditionally, the motor cortex (MCtx), the basal ganglia (BG), and the cerebellum (CB) have been considered key motor control regions of the brain, and plasticity within these regions are known to support motor learning. In addition, neuromodulation, such as from adrenergic neurons in the locus coeruleus (LC), is critical for proper behavior and learning. Despite the key role these regions play in controlling movements and their implication in movement disorders, we are only beginning to understand how motor signals from these regions interact with each other. Anatomically, the thalamus serves as a common target structure for MCtx, BG, and CB, yet conventionally, the thalamus has been viewed as a passive relay station. However, there is emerging evidence that the thalamus can functionally integrate and modulate these diverse motor signals. How thalamic neurons respond to motor inputs, the role of motor thalamus in motor learning, and how adrenergic signaling modulates thalamic activity are largely undefined. My central hypothesis is that the motor thalamus serves as a point of convergence for motor signals from MCtx, BG, and CB, as well as neuromodulatory input from LC, allowing it to functionally integrate these inputs to control movements and promote motor learning. I propose to use a combination of in vivo deep-brain imaging and novel fluorescent sensors for intracellular signaling in mice performing motor tasks, as well as slice electrophysiology, to measure the activity of motor thalamus during movement and determine how such activity is modulated by adrenergic input from LC. These approaches will allow me to define the inputs to motor thalamus and measure thalamic activity during movement and motor learning (Aim 1), determine the functional role of motor thalamus and its inputs in motor control (Aim 2), and determine how adrenergic signaling modulates thalamic activity during motor learning (Aim 3). Results from this study will not only clarify the role of the motor thalamus in motor control and motor learning but also provide an understanding of how adrenergic neuromodulation influences thalamic activity during behavior. This is of critical importance, as abnormal thalamic activity and disrupted adrenergic signaling are characteristic features of motor diseases. The experiments proposed in this study will span the mentored K99 and independent R00 phase of this award, with the K99 phase being focused on defining the activity and role of motor thalamus during motor learning and the R00 phase focused on understanding how adrenergic input to the thalamus modulates motor signals. My proposed training plan builds on my experience in two-photon in vivo imaging and mouse behaviors and will add training in slice electrophysiology. In addition, my expert mentoring team will also provide guidance in my career development with the goal of launching a successful career as independent investigator at a research institution.

NIH Research Projects · FY 2024 · 2023-08

Accumulating research suggests that barriers to eliminating the persistent disparities in cardiovascular disease (CVD) are related to structural-level social determinants of health (SDOH). The majority of this evidence is cross-sectional, from studies using administrative datasets (i.e., US Census) to quantify structural SDOH associations with ecological-level measures of CVD. Prospective and clinical CVD outcome data are needed to advance from descriptive-level evidence; however, well-established cohort studies typically lack access to novel structural determinants. The scientific objective of the research plan is an innovative solution to generate the needed high-quality dataset, by employing data fusion techniques to link structural determinants from administrative datasets with prospective cohort data. I will generate four structural-level determinants at the neighborhood-level using geographic linkages between the Women’s Health Initiative (WHI) cohort with 1) US Census 2) American Community Survey (ACS) 3) Center for Disease Control and 4) Neighborhood Redlining Maps. Each structural determinant adheres to recent conceptual frameworks for advancing the quantification of CVD disparities. I uniquely measure determinants longitudinally to account for changes in residence and the duration of exposure. In Aim 1 (K99 phase), I will quantify structural determinants at the intersection of demographic and material determinants using the index of concentration at the extremes (ICE). The causal effects of ICE on CVD incidence over 30 years of follow-up will be estimated. In Aim 2, I propose to link the Social Vulnerability Index to evaluate a hypothesized structural intervention on CVD. In Aim 3, I propose to estimate CVD risk associated with demographic residential segregation and residence in a historically redlined neighborhood. Evaluating causal mechanisms, temporality, life-course exposure, and accounting for intersectionality would markedly advance the current level of evidence. The public health implications of which may help design future interventions to target modifiable structural policies and practices. The career development plan will advance my scientific training in data fusion techniques, the modeling of structural determinants, and pathways to CVD. Through mentored training combined with this research plan, the K99/R00 will prepare my transition to an independent investigator in a tenure-track faculty position. This award would advance three Objectives of the NHLBI Strategic Vision through the use of (3) an emerging opportunity in data science to accelerate understanding of (7) factors that account for differences in health among populations, led by (8) a scientist who would “further develop, [redacted], and sustain a scientific workforce capable of accomplishing the NHLBI’s mission”.

NIH Research Projects · FY 2025 · 2023-08

Overall Modified Project Summary/Abstract Section Postpartum hemorrhage (PPH) is a leading cause of maternal death and severe maternal morbidity (SMM) with disproportionate effects across a population. We propose to create a Maternal Health Research Center of Excellence at Stanford University called PRIHSM (PRomoting Improvement in Hemorrhage-related Severe Maternal morbidity). The goals of PRIHSM are to reduce PPH and associated SMM by addressing important precursors to PPH-related SMM and mortality: iron deficiency anemia (IDA) and obstetric care variability contributing to cesarean birth. The problem of maternal IDA is vastly under-appreciated yet it affects approximately 16% of pregnancies in the U.S. We propose that by effectively addressing antenatal IDA, we can reduce PPH-related SMM. Additionally, almost 1 in 3 U.S. births are by cesarean delivery and rates vary 10-fold across hospitals. While cesarean section can be a lifesaving intervention when appropriate, it is associated with significant risks that include PPH. We propose that addressing variability in obstetric care is an important strategy in reducing cesarean-linked PPH and reducing PPH-related SMM. Thus, our Aims are to: Aim 1 (Project 1). Reduce antenatal IDA by developing, implementing, and disseminating a patient-informed Anemia Prevention Toolkit, which will standardize the evaluation, diagnosis, and treatment of IDA and reduce the prevalence of IDA at birth admission and PPH-associated SMM. Aim 2 (Project 2). Reduce rates of primary cesarean birth and cesarean-linked PPH by conducting a mixed methods study to understand obstetric care-related drivers of hospital-level rates of these outcomes, and implementing a patient-informed Hospital Action Guide. Our work will involve community-university partnerships focused on improving maternal health and be driven by perspectives of patients, providers, and healthcare leadership. Our work will provide training opportunities to build research and clinical expertise relevant to PPH, including individuals who represent experiences within academic and community-based settings, and underserved areas. We propose a bold yet achievable agenda that will affect a sustainable decline in PPH-related mortality and morbidity.

NIH Research Projects · FY 2025 · 2023-08

Impaired postpartum sleep significantly impacts postpartum recovery, maternal health, and the maternal-infant dyad. Little is known about postpartum sleep as a construct, postpartum sleep disorders, or their association with mental health. Data are limited to small non-racially diverse cohorts using measures with inadequate content validity. Our pilot work demonstrates that one-third of patients with a postpartum sleep disorder have co-morbid postnatal depression, a leading cause of maternal suicide. We propose to test our overall hypothesis that postpartum sleep is inadequately assessed, and postpartum sleep disorders are underdiagnosed and linked to postnatal depression. The Structured Clinical Interview for Sleep Disorders (SCISD-R) diagnoses common DSM- 5 sleep disorders, including insomnia disorder. However, SCISD-R is impractical outside research settings. Wrist-worn actigraphy devices can also be used to measure sleep objectively. In contrast, patient-reported outcome measures (PROs) are inexpensive, self-reported tools, which can be used to assess postpartum well- being in large cohorts. Our recent work highlights that existing sleep PROs inadequately assess important postpartum sleep domains such as infant sleep, nocturnal feeding patterns, awakenings, and protected maternal sleep time. Clinicians currently lack robust PROs to assess postpartum sleep and screen for postpartum sleep disorders. We therefore propose developing, calibrating, and validating sensitive, context-specific, sleep PROs that capture the symptoms and dimensions of postpartum sleep, while screening for postpartum sleep disorders. To achieve this objective, we will implement novel Patient-Reported Outcomes Measurement Information System® (PROMIS®) ‘gap’ methodology to: (1) adapt existing PROMIS Sleep Disturbance and Sleep-Related Impairment PROs for postpartum use by identifying new PRO items, which improve their content validity, through qualitative interviews and focus group meetings with ≥75 stakeholders and cognitive debriefing interviews with ≥20 patients; (2) calibrate and finalize the adapted postpartum sleep PROs through psychometric evaluation in a racially and socioeconomically diverse cohort of 640 (160 White non-Hispanic, White Hispanic, Black and Asian) women recruited across two hospitals in Stanford, CA and Little Rock, AR. Women will complete sleep, global health and Edinburgh Postnatal Depression Scale PROs at 4 postpartum time points (inpatient, 3 and 6 weeks, 6 months). SCISD-R will also be conducted in 300 of these women at 6 weeks and 6 months, as a gold standard, to determine screening sensitivity and specificity of the adapted PROs. In 50 women, sleep data, (day, nocturnal and total awake time), will also be compared between actigraphy and the adapted PROs; and (3) characterize postpartum sleep and its association with postnatal depression using causal inference approaches. Rigorous validation of the sleep PROs will support their widespread use to assess postpartum sleep, screen for sleep disorders, identify disparities, explore links with postnatal depression, and to promote maternal health and well-being. As such the proposed study aligns well with Goals 3 and 4 of the 2021 NIH Sleep Research Plan.

NIH Research Projects · FY 2026 · 2023-08

ABSTRACT Dr. Annesa Flentje is Associate Professor in Community Health Systems and Department of Psychiatry and Behavioral Sciences at the University of California, San Francisco. She mentors early career researchers in sexual and gender minority (SGM, non-heterosexual and transgender or gender non-binary people, respectively) health, focusing on substance use, minority stress, and epigenomic markers of substance use. SGM people have greater substance use when compared to their cisgender, heterosexual counterparts. The greater rates of substance use are attributed to minority stress exposure (unique stress burden due to discrimination and stigma among SGM people), and substance use increases in the presence of minority stress. Further, unique molecular profiles of both substance use and minority stress have been observed in sexual minority men living with HIV, suggesting that alterations in the epigenome may serve as biological markers for substance use. Unfortunately, research to date is limited because these models have not explicitly compared people living with HIV (PLWH) and people living without HIV (PLWoH). Further, SGM people have unique hormonal exposures that have been unaccounted for in research investigating substance use and potential epigenetic biomarkers for substance use and these hormonal exposures may also be related to alterations in the epigenome. This project will expand Dr. Flentje’s research program to integrate hormonal exposures and HIV status, to be able to identify epigenomic markers of substance use in the presence of endogenous and exogenous hormone exposures comparing PLWH to PLWoH. To support expansion of her research, Dr. Flentje will receive additional training in HIV, hormone exposures, dominance analysis, and epigenetic bioinformatics analysis. This K24 will support Dr. Flentje in mentoring patient-oriented researchers in SGM health who will leverage survey and epigenetic data from existing cohort studies: The PRIDE Study, the MACS/WIHS Combined Cohort Study, and the All of Us Research Program to 1) understand key minority stress predictors of substance use among SGM people and compare the relative strength of these predictors between PLWH and PLWoH; 2) identify endogenous and exogenous hormonal predictors of substance use among SGM people and determine differences in the relative strength of these predictors among PLWH and PLWoH; and 3) derive minority stress, substance use, and hormonal phenotypes among SGM people, identify epigenetic markers of these phenotypes, and identify differences in these epigenetic markers between PLWH and PLWoH. This K24 will support mentorship of early career researchers in SGM health focusing on substance use, HIV, hormonal biology, epigenomics, and minority stress. It will also expand Dr. Flentje’s mentorship skills to integrate structures to support mentees in navigating financial challenges, loan repayment applications, family building, and emotional hardships. As substance use research among SGM people is an emerging area of study, a national approach to mentorship to promote innovation in patient-centered research is critical.

NIH Research Projects · FY 2024 · 2023-08

Project Summary Sleep is a critical behavioral state that fulfills essential needs for health, including clearing waste products (e.g., amyloid beta [Aβ]) from the brain. As humans age, sleep quality strikingly deteriorates, and this decline correlates with increased risk for neurodegeneration, vascular dementia, and Alzheimer’s disease. While the occurrence of sleep disruption during aging is well documented, the causative impact of sleep on brain resilience with age and disease remains unexplored. I hypothesize that sleep is a key modulator of animal health that can be manipulated to improve brain resilience in the context of aging and disease. To investigate the impact of sleep on brain resilience late in life, I will (Aim 1) characterize if age-associated sleep deterioration (e.g., circadian timing and amount of sleep) impacts cognitive health, (Aim 2) perturb sleep and test the impact on cognitive resilience late in life, and (Aim 3) determine if sleep improves brain resilience in the context of human Aβ1-42 overexpression. The age dependence of sleep deterioration and neurodegeneration is difficult to study at scale due to the time-consuming challenge of aging vertebrates. To overcome this challenge and tackle this question, I will use the African killifish, a model with an extremely short lifespan of only 4-7 months. The killifish exhibits key hallmarks of human aging (e.g., neurodegeneration, frailty) and has conserved brain structures and genes known to regulate sleep. Critically, killifish brains exhibit increases in neurofibrillary degeneration, oxidative stress, gliosis, and inflammation, as well as decreases in repair, as they age. The killifish also possesses practical advantages such as low husbandry costs, a short generation time (<1 month), and genetic tractability. These traits make the killifish a suitable model system to investigate how sleep may impact brain resilience with age. In preliminary efforts, I built a longitudinal tracking system to generate an unprecedented view into how sleep changes across the lifespan, and I found that killifish exhibit an age-associated sleep decline that parallels human sleep decline. I also genetically perturbed sleep and identified novel lifespan-extending genes. I used my new CRISPR knockin method to develop the first killifish model for Alzheimer’s disease. Using these tools and discoveries, I will determine how sleep impacts brain resilience with age and disease. I am pursuing this project at Stanford University with training from my mentor Dr. Anne Brunet, co-mentor Dr. Karl Deisseroth, and an exceptional scientific advisory team whose expertise spans brain aging, Alzheimer’s disease, neurodegeneration, and sleep. Through continued training with the K99/R00 award, I will learn new methods (killifish genetics, intact whole-mount brain staining, and advanced transcriptomic/behavioral data analysis) and concepts (the biology of aging, Alzheimer’s disease, protein aggregation, neurodegeneration). This work, my career development, and my technical training will provide me with the skills and knowledge required to be a successful leader of a laboratory at a top academic institution.

NIH Research Projects · FY 2024 · 2023-08

Project Summary We propose to contribute a Bay Area, population-based prospective cohort to a nationwide cohort study of Asian American (AsA) men and women to address major gaps in evidence on the correlates and determinants of disease risk and health. We will recruit, characterize, and follow Chinese, Filipino, and Vietnamese Americans, three understudied AsA ethnic groups whose populations are rapidly expanding in the United States. Specifically, we will recruit 2,100 Chinese, Filipino, and Vietnamese Americans (700 of each ethnic group) in the San Francisco Bay Area to characterize cardiometabolic profiles and psychosocial and other health factors through extensive surveys, clinical assessments and assays, imaging studies, multi-omics, and digital technology. Although disease patterns and risk factors appear to vary among different AsA groups, previous research suggests that AsAs are highly affected by cardiometabolic disorders such as diabetes, hypertension, dyslipidemia, cardiovascular disease, stroke, non-alcoholic fatty liver disease, and obesity, as well as underdiagnosed mental health conditions and psychosocial issues related to immigration trauma, discrimination, and marginalization. However, epidemiological data on individual AsA groups are sparse, and most clinical guidelines and treatments are based on data derived from Caucasians. Our UG3/UH3 study site will collaborate with the Coordinating Center (U24), other study sites, and NHLBI to develop a large cohort of 10,000 members to characterize risk factors and disease patterns in individual AsA ethnic groups. The specific aims of the proposed study are to 1) contribute to the development of a Common Protocol for the study in collaboration with NHLBI, U24, and other UG3/UH3 investigators and establish a state-of-the-art populomics cohort; 2) measure the prevalence or distribution of baseline self-reported health and risk factors and clinical markers in each ethnic group and compare across ethnic groups; and 3) determine relationships among baseline risk factors in each AsA ethnic group, including self-reported stress and sleep as well as heart rate variability-derived sleep quality and stress from wearable biosensing data, with subsequent (incident) health outcomes during follow-up. This proposed study is innovative in its in-depth exposome phenotyping through comprehensive surveys that will include assessments of well-being and anxiety in addition to traditional risk factors, baseline clinical assessments, advanced digital technology, and integrative omics to understand the intersections of epidemiology, biology, psychology, and technology in physical and mental health. We will leverage these innovations alongside Stanford’s world-class resources in population health sciences to uncover critical cardiometabolic and psychosocial factors underlying health and disease in AsAs. This study will advance health and disease prevention and treatment and will lead to improved health and well-being outcomes for AsAs. 1

NIH Research Projects · FY 2025 · 2023-08

We propose to contribute a San Francisco Bay Area (Bay Area) population-based prospective cohort to a nationwide cohort study of Asian Americans (AsAs) to address major gaps in evidence on the correlates and determinants of disease risk and health. We will recruit, characterize, and follow Chinese, Filipino, Indian, and Vietnamese Americans, four understudied but high-risk AsA groups. These communities are expanding rapidly in the United States (US), making them a high priority to enhance overall population health. Specifically, we will recruit 2,100 AsAs in the Bay Area to characterize cardiometabolic profiles and psychosocial and other health factors through extensive surveys, clinical assessments and assays, and imaging studies. Although disease patterns and risk factors appear to vary among different AsA country-of-origin groups, previous research suggests that AsAs are at high risk of cardiometabolic disorders such as diabetes, hypertension, dyslipidemia, cardiovascular disease, stroke, non-alcoholic fatty liver disease, cardiovascular-kidney-metabolic syndrome, and obesity, as well as underdiagnosed mental health conditions. However, epidemiological data on individual AsA groups are sparse, and clinical guidelines and treatments specific to these groups have not been developed. Our UG3/UH3 study site will collaborate with the Coordinating Center (U24), other study sites, and NHLBI to develop a large cohort of 10,000 people to characterize risk factors and disease patterns in individual AsA groups. The specific aims of the proposed study are to 1) contribute to the development of a Common Protocol for the study in collaboration with NHLBI, U24, and other UG3/UH3 investigators and establish a state-of-the-art populomics cohort; 2) measure the prevalence and distribution of baseline clinical markers and self-reported health and risk factors in each AsA group and compare across groups; and 3) determine relationships among baseline risk factors in each AsA group, including clinical data and self-reported stress and sleep, with subsequent (incident) health outcomes during follow-up. This proposed study is innovative in its in-depth exposome phenotyping through comprehensive surveys that will include assessments of mental health and traditional risk factors, in addition to in-depth clinical data and biospecimens, to understand the intersections of epidemiology, biology, psychology, and technology in physical and mental health. We will leverage these innovations alongside Stanford’s world-class resources in population health sciences to uncover critical cardiometabolic and psychosocial factors underlying health and disease in AsAs. This specific cohort development is for a high-risk group and is related to a larger population-based effort to address health disparities.

NIH Research Projects · FY 2025 · 2023-08

Abstract Stroke is a disabling cerebrovascular disease that causes 5.5 million deaths each year globally. The disease progresses rapidly and irreversibly, leaving a narrow time window for intervention. Existing methods for patient selection for endo- vascular thrombectomy are suboptimal, based exclusively on simple linear threshold models applied to neuroimaging. Deep learning has shown great promise in recent years for many medical applications. We believe that it can be used to integrate imaging and non-imaging data in a seamless and data- driven way to improve stroke triage and clinical trials. The goal of this project is to develop deep convolutional neural network approaches to the initial MR and CT imaging, the most commonly performed stroke imaging protocol in acute ischemic stroke patients, and to combine this with non-imaging clinical information. We will train networks to predict the most likely final tissue and clinical outcomes under 2 extreme conditions (major reperfusion and minimal reperfusion) to estimate the treatment effect at the individual level. Next, we use the methods and learning from this first study to train deep learning models without using contrast perfusion imaging, which will improve safety, cost, and time-to-treatment. Finally, we will test the generalizability and explainability of these AI methods in external cohorts which differ in terms of population and scanner types, including testing on data from mobile CT scanners. Accomplishment of these aims will fundamentally shift the acute stroke paradigm beyond the relatively simplistic mismatch concept and replace it with a data-driven method that takes into account the immense amount of imaging and clinical data that can be brought to the stroke decision-making process. The methods developed will improve long-term outcomes and reduce of the cost of stroke care worldwide.

NIH Research Projects · FY 2024 · 2023-08

PROJECT SUMMARY Understanding the general rules of adaptation has implications for treating diseases caused by evolving entities such as cancer and drug-resistant infections. In attempts to uncover these general rules, previous research has revealed a seemingly paradoxical phenomenon. On the one hand, under selective pressure, adaptive mutations arise quickly, have large fitness effects, localize to a few functionally similar genes, and likely exploit the same adaptive strategy. On the other hand, laboratory-evolved organisms show a wide range of diverse physiological changes. This means that directional selection increases rather than decreases phenotypic diversity. This phenomenon likely explains our limited ability to predict the effectiveness of anticancer therapies, because even genetically similar cancers can have highly diverse physiologies. In this project, I will use a collection of several hundred well-studied adaptive S. cerevisiae mutants, isolated from a single evolution experiment, to test two mechanistic hypotheses of how selection might generate diversity. My working hypothesis is that selection generates diversity through combinatorial loss of plastic responses. The alternative hypothesis is that selection generates diversity through the gain of novel responses. To distinguish between these two hypotheses, I will measure the molecular phenotypes of the adaptive mutants with two RNA sequencing technologies. In Aim 1, I will use bulk RNA sequencing on 10 carefully selected mutants, grown in the environment in which they evolved, in order to gain the first insight into the molecular mechanism of their adaptive strategy. In Aim 2, I will use a state-of-the-art, high-throughput, barcode-aware RNA sequencing approach called Split-Seq to measure the full scope of the adaptive mutants' phenotypes which are most prominent in various extreme environments. This will reveal if the adaptive mutants achieve high fitness in the evolution environment and high diversity in other environments through loss of plasticity or by creating novel responses. Uncovering the mechanism of how selection generates diversity will contribute to our general understanding of adaptation and have implications for treating diseases driven by evolutionary adaptation, such as cancer and drug-resistant infections. During my PhD, I received strong training in molecular genetics and learned basic wet lab skills but had little exposure to evolutionary theory or genomics. Undertaking this project will greatly expand my conceptual and technological toolkit, as I will learn molecular evolution and population genetics, as well as genomics methods like barcode tracking and RNA sequencing. I will receive formal and informal training in evolutionary theory, programming, genomics data analysis, grant writing, and professional leadership. This will prepare me to become an independent researcher and leader in the interdisciplinary field of molecular and evolutionary biology.

NIH Research Projects · FY 2026 · 2023-08

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 safe and effective in animal models. It is currently under active investigation in humans. However, a major concern in the use of TUS in human studies is the potential for an off-target auditory stimulation by bone-conducted sound, making the study susceptible to placebo and attention effects, which are relevant in most studies of neurological disorder interventions. The purpose of this proposed research is to increase the rigor and safety, and therefore success, of TUS in the treatment of SUDs, and TUS more generally, thereby enabling truly innovative science. We will do this by accurately quantitating the hearing response from TUS waveforms in humans, including correlation to skull morphology, and designing and testing a framework for creating waveforms that reduce audibility while improving masking, all while still providing the flexibility needed by TUS researchers. We will do this for stimulation waveforms and for MR-ARFI waveforms, which we show in preclinical data provide a measure of the dose-response to the ultrasound stimulation.