Northeastern University

NIH Research Projects · FY 2026 · 2023-12

Summary Injuries to the growth plate (GP) remain a major cause of morbidity in children, often leading to angular limb deformities and even complete growth arrest. Slipped Capital Femoral Epiphysis (SCFE), a transverse Salter-Harris type 1 (SH1) fracture, where the spherical femoral head of the hip joint slips backwards off the femoral neck, is a common and debilitating disorder, affecting the hip in children and adolescents and often culminating in painful arthritis in adults. A vertical Salter Harris type 2-4 (SH2-4) fracture, in which the fissure plane extends in the metaphysis or epiphysis, leads to an osseous bridge formation in lieu of cartilage, and angular deformities. This highlights the need for novel therapeutic strategies to prevent bone formation at the injury site and to allow epiphyseal cartilage to resume growth. One avenue to promote local cartilage repair, would be through mobilization of resident stem cells in response to injury. Current literature describes the GP-RZ as a stem cell-rich region, which give rise to the GP appositionally. We recently discovered that a population of FoxA2+ long-term stem cells (LTSSC) is located in the top compartment of the RZ. FoxA2+LTSSC have higher longevity, clonogenicity, and are mutually exclusive with the previously characterized PTHrP+ short-term stem cells (STSSC), located at the bottom of the RZ. A horizontal SH1 injury successfully heals with hyaline cartilage and involves the expansion of FoxA2+ LTSSC. A large vertical SH4 fracture heals through a "bony bar", while a small SH4 lesion heals with hyaline cartilage. FoxA2+ LTSSC line the outside of a large SH4 wound, but infiltrate the small SH4 lesion. This suggests other cell types (e.g osteoprogenitors) may create a physical barrier or a signaling niche, which hinders FoxA2+ LTSSC healing of large defects. In both injuries, a transient pulse of TGFβ signaling, induced in cartilage remnants at D1, precedes FoxA2+ LTSSC expansion at D3. This indicates a potential role for TGFβ, in promoting FoxA2+ LTSSC expansion and differentiation into chondrogenic progeny, for cartilage repair. These findings will allow us to investigate, for the first time, whether FoxA2+LTSSC contribute to GP cartilage repair following SH1 and SH4 injuries. To test this premise, we propose three aims. In Aim 1, we will determine if FoxA2+ cells can contribute to the repair of a small SH4 lesion, but fail to contribute to the repair of a large SH4 lesion. In Aim 2, we will determine if TGFβ signaling drives FoxA2+LTSSC migration and differentiation into progeny to accommodate cartilage repair after SH1 injury. In Aim 3, we will determine if FoxA2+cells transplantation in a large SH4 lesion can prevent formation of a bony bar. Altogether, these results will provide a strong basis for development of bioengineering strategies for GP cartilage regeneration.

NIH Research Projects · FY 2026 · 2023-11

Project Summary The bulk of increased resistance to the outflow of aqueous humor in glaucoma occurs at the vicinity of the inner wall endothelium of the Schlemm’s Canal (SC). The giant vacuoles (GVs) and pores associated with the SC endothelial cells are the only open spaces for the aqueous humor to enter the canal. Thus, they are thought to play an important role in regulating outflow resistance. GVs form in response to a basal to apical pressure gradient that subjects SC cells to substantial deformation. The extent of this deformation is mediated by SC cell mechanics. We recently discovered that the elevated outflow resistance in glaucomatous human eyes is associated with the increased stiffness of their SC cells in situ. These observations render SC cell stiffness a key factor in GV formation and outflow homeostasis. Yet, little is known about the mechanism(s) that regulate the biomechanical properties of SC cells and GVs. We previously showed that SC cells become stiffer when cultured on stiffer substrates in vitro. We also recently showed that glaucomatous SC cells and their underlying extracellular matrix are stiffer than their healthy counterparts in situ. These findings suggest that the mechanical properties of SC cells are substrate dependent. In this project, I aim to examine the role of the vimentin intermediate filament (VIF) cytoskeleton in regulating the biomechanical properties of SC cells and their associated GVs. The reasons for my interest in VIFs are twofold: first, they are shown to be major contributors to cell mechanics in general, they are the dominant determinant of cell mechanics at large deformations, and they have a substrate stiffness dependent assembly state; secondly, VIFs are highly expressed in SC cells, and it has been shown that they associate with GVs in situ and also impact their life cycle in vitro. To examine this role, I will first knockdown vimentin in cultured human SC cells and use atomic force microscopy and traction force microscopy to establish the role of VIFs in SC cell stiffness and contractility. The findings from these studies will be used as a basis for additional studies employing super-resolution imaging, biochemistry, and microfabrication to determine how substrate dependent expression and assembly states of VIFs in SC cells affects their stiffness and contractility. I will next investigate the role of VIFs in GV formation through ex vivo perfusion of eyes from wildtype and vimentin knockout mice followed by characterizing and comparing the GV size and density along their SC. I will also determine the impact of the presence or absence of VIFs on the generation of outflow resistance by measuring the outflow facility in these eyes. Finally, I will knockdown vimentin in normotensive mouse inner wall to determine the feasibility of targeting VIFs for modulating outflow facility. I will then extend this method to ocular hypertensive mice to gauge the effectiveness of this approach as a novel treatment for glaucoma. Through examining the contribution of VIFs to the biomechanics of SC inner wall, I seek to transition into an independent career in order to investigate the mechanical basis of increased outflow resistance in glaucoma and to develop novel therapeutic approaches for the disease.

NIH Research Projects · FY 2025 · 2023-09

ABSTRACT Mucins and other densely glycosylated proteins play critical roles in a number of biological processes, disease conditions, and therapeutics. The functioning of these sugar-coated molecular machines depends on their structure, dynamics, and conformational transitions. Experimental techniques for capturing such structural dynamics, however, can be extremely challenging and resource intensive. We seek to improve upon some of the existing glycan modeling computational tools as well as design new in silico techniques, as robust alternatives to experimental studies. These tools will be used to build interconnected mucin glycoprotein gel systems with native glycosylation patterns, and obtain understanding of functional underpinnings at the molecular level. Effects of perturbations in terms of pH variance, varying glycosylation patterns, and charge distribution changes will be investigated. This will enable detailed comprehension of the physical properties of mucins that drive their function, as well as the molecular elucidation of disease conditions of cystic fibrosis, mucosal inflammation, and mucin-mediated cancers. A multi-modal approach will be employed to study these mucin networks in different scales – (i) first-principles based atomistic modeling to capture the equilibrium structure-dynamics; (ii) biophysics-based coarse-grained methods to describe bulk properties and transitions, and (iii) data-driven machine learning approaches to predict topology and intermolecular interactions. Inspired from mucosal gels, we will use these tools to design novel mucin-like nanomaterials constructed from glycan-peptide heteropolymer networks to target different biomedical applications. We aim to optimize a machine learning (ML)-driven combinatorics method for glycan arrangement in these polymers that will provide enhanced control over material properties – a molecular LEGO of glycans geared towards customizable mucin-mimetic biomaterials.

NIH Research Projects · FY 2025 · 2023-09

Project Summary/Abstract Organ surfaces are covered with epithelial cells or endothelial cells, providing physical barriers for organs and bodies. Cells on these confluent layers often remain static and non-migratory. However, they can also undergo active structural rearrangements during basic physiological processes ranging across embryonic development, morphogenesis, repair, and remodeling. In each of these events, an epithelial collective necessarily undergoes a transition from a solid-like state which is quiescent and non-migratory to a fluid-like state which is dynamic and migratory. This striking transition between non-migratory versus migratory behaviors is traditionally studied in the context of cells on a flat surface in 2D. These collective cellular behaviors have been widely explored in monolayers of epithelial cells that form two-dimensional (2D) flat surfaces, from both biophysics and cell biology perspectives. However, they are not well-adapted to make predictions for natural epithelia, which are typically found to form highly curved surfaces, where the radius of curvature can be comparable to a few cell lengths. Epithelial tissues also comprise various topologies – spheres, ellipsoids, tubes, and saddle points — in native structures such as embryos, alveoli, airways, vessels, and branching bifurcations. How surface curvature affects the way a cell collective moves remains largely unknown; furthermore, how cells become jammed and unjammed during the maturation of a cell monolayer growing on a curved surface remains unclear. Further, whereas previous modeling efforts have focused more on the mechanics and migratory behavior of cells within a single monolayer, the mammalian epidermis is a multilayered epithelial tissue. Although the developing epidermis is highly dynamic, the time-dependent mechanics (i.e., rheology) of epidermal development remains elusive. There are two key unresolved questions: (1) what cues drive epidermal development, and (2) how does the mechanics of the epidermis depend on the timescale of measurement? There is an urgent need to develop theoretical and computational models for these critical scenarios. I will develop an integrated computation modeling framework to elucidate the biomechanics of collective cell behavior beyond the conventionally studied two-dimensional settings, including curved surfaces and multilayered 3D epidermis. I will also create a novel model that addresses the biomechanical couplings between nuclear morphologies and epithelial proliferation.

NIH Research Projects · FY 2026 · 2023-09

PROJECT SUMMARY/ABSTRACT Advancing Medical Illustration in Patient Education Materials: from Art to Science Despite the enormous effort and expense spent in the creation of medical illustrations, there has been no systematic empirical evaluation to date of the impact of varying illustration elements and styles on patient comprehension of the concepts that were intended to be conveyed in the illustration or patient anxiety induced by the illustration. In this effort we will first develop a domain ontology that systematically describes the space of medical illustration styles in current use. We will then use this ontology to guide the creation of a public domain corpus of patient education documents that exhibit the most important illustration distinctions identified in the ontology, along with knowledge tests for each. This corpus will be used to conduct an evaluation involving 8,100 individuals to assess the impact of different illustration styles on comprehension and anxiety. Finally, we will explore the use of Embodied Conversational Agents that simulate face-to-face conversation with a health provider, to explain patient education documents, both on 2D displays and in immersive Virtual Reality, evaluated in a randomized study involving 300 participants.

NIH Research Projects · FY 2024 · 2023-09

ABSTRACT Oral modified-release (MR) drug products with modulated drug release characteristics (e.g., rate, duration, and the site of drug release) have been widely used to achieve desired therapeutic effects, reduced adverse effects and/or improved patient compliance than conventional oral solid dosage forms. More than half of the FDA approved oral MR drug products are extended-release (ER) tablet products with multiple strengths. Until now, appropriate factors to scale the formulation for additional strengths for oral MR tablets have yet to be determined. Moreover, the key variables affecting drug release mechanism and formulation design spaces for different MR technologies have not been fully understood and identified. The main objectives of this project are to: 1) determine the impact of formulation variables (e.g., drug properties and excipients) on the drug release mechanism of in-house made ER tablets based on quality-by-design (QbD) principles; 2) develop mechanistic models parameterized with dissolution data obtained using comprehensive dissolution testing technologies to compare the ER tablets and the corresponding reference drug products across multiple strengths to establish dissolution safe spaces and to identify critical quality attributes (CQA’s); and 3) construct a “proof-of-concept” machine learning model leveraging the database of complex oral MR drug products to identify key variables that affect drug release mechanisms for different formulation design strategies. The proposed research builds upon our extensive research on the formulation development, comparative product characterization, in vitro dissolution testing as well as bioequivalence assessment and mechanistic modeling of complex oral MR solid dosage forms. Biopharmaceutics classification system (BCS) Class I and Class II compounds ropinirole hydrochloride and nifedipine will be studied as model drugs, respectively. ER tablets across multiple strengths with formulation and process variables will be produced and comparatively characterized using the corresponding reference drug products as controls. The drug release mechanism and in vitro dissolution profiles of the ER tablets across multiple strengths will be characterized under different testing conditions including fasted and fed conditions with simulated gastrointestinal motility. Moreover, mechanistic models (e.g., physiologically based pharmacokinetic (PBPK), physiologically based biopharmaceutics models (PBBM)-PBPK) parameterized with the in vitro data obtained will be developed to identify appropriate factors to scale the formulation for additional strengths for oral MR tablets. Lastly, a comprehensive database of the approved oral MR drug products will be established, and ML techniques will be employed to identify the key variables that impact drug release mechanism. The proposed research will help advance the regulatory review and approval processes of oral MR tablet products, and support the approval of additional strengths for such drug products. Facilitating the development of complex generic oral MR drug products will ultimately help increase the public access to high quality and affordable oral medications.

NIH Research Projects · FY 2024 · 2023-09

Abstract Yoga, a physical activity based mind-body therapy has become increasingly popular as a preventive and therapeutic practice, making it one of the therapies with the most rapid increase in prevalence. Practice of yoga combines physical activity, modulated breathing and meditative exercises that combine to provide the practitioner a holistic mind-body experience. The effectiveness of yoga has been examined for a number of health outcomes, including stress, anxiety, depression, management of chronic conditions such as lower back pain and type 2 diabetes. Fewer studies have examined the effects of yoga practice on brain health and cognitive function. Age related cognitive decline is well documented. With the increasing population of older adults in the US, there is a need to not only test popular forms of exercise such as yoga, but also compare their efficacy with modalities such as aerobic exercise that have proven neurocognitive benefits. This research is timely in that yoga is a gentle and modifiable form of exercise that can be easily adapted for practice by older adults who may exhibit comorbidities or chronic health conditions that keep them from walking or performing other forms of aerobic exercise. To our knowledge, this is the first randomized control trial examining and comparing the effects of yoga and aerobic exercise training on behavioral and neurobiological correlates of cognitive function. We propose to conduct a 6-month randomized controlled exercise trial among older adults to compare the efficacy of yoga with aerobic exercise on cognitive function, brain structure and function, cardiorespiratory fitness, functional fitness, and inflammatory and molecular markers. Using a single-blind, three arm randomized control trail, 168 older adults ages 55-79 will be assigned to either: a Hatha yoga group, an aerobic exercise group or an active stretching and strengthening control group. The groups will engage in hour-long group exercise sessions 3x/week. A comprehensive neurocognitive test battery, brain imaging, cardiovascular fitness test, and a blood draw will take place at baseline; end of the 6-month intervention, and at 12-month follow-up. The proposed pilot RCT will examine the relationship between yoga training and improved cognitive functioning as well as identify neurobiological correlates as potential mechanisms of action through which yoga training exerts its effect on cognitive function. The investigative team is interdisciplinary, highly productive and has a history of collaboration in conducting exercise interventions examining neurocognitive outcomes in older adults.

NIH Research Projects · FY 2024 · 2023-09

Project Summary The overarching goal of this study is to measure the prevalence of cancer misinformation on social media and understand the mechanisms that underlies its spread. Belief in misinformation can have serious ramifications, particularly when the misinformation is regarding life threatening conditions such as cancer. We currently lack answers to even basic questions regarding cancer misinformation online. For example, how much cancer misinformation is there on social media? How do people make assessments of trust and source credibility? How well do people update their beliefs when cancer misinformation is corrected? What are the psychological mechanisms of this belief updating? This Pathway to Independence Award (K99/R00) application by Dr. Briony Swire-Thompson intends to fill this knowledge gap by building on her prior research regarding misinformation prevalence on social media, source credibility, and the correction of misinformation. The proposed research will be complemented by focused training on three areas, (1) increasing Dr. Swire- Thompson’s knowledge of cancer and cancer misconceptions (2) furthering her social media data skills, and (3) fostering professional development to facilitate the transition into an independent research position. These training goals will be supervised by an interdisciplinary mentoring team. This team will be led by Dr. Lazer, a University Distinguished Professor of Computer Sciences at Northeastern University’s Network Science Institute. The co-mentors will be Dr. Viswanath, a cancer communications expert and Lee Kum Kee Professor of Health Communication at the Harvard T. H. Chan School of Public Health, and Dr. Johnson, an oncologist and Assistant Professor at the University of Utah’s School of Medicine. This training will aid Dr. Swire- Thompson to answer three specific research aims. First, the prevalence of cancer misinformation on social media will be investigated. In the K99 phase, she will focus on Twitter to investigate who is more likely to be exposed to and share cancer-related misinformation. In the R00 phase this will be extended to Facebook, where we invite a representative sample of individuals to donate their social media data and respond to surveys regarding their relationship with cancer, and why they share information. The second specific aim is to investigate how people make judgements of source credibility, and the extent that credibility is reduced when cancer sources are disreputable (such as spreading misinformation or having a lack of expertise). The third specific aim is to understand the cognitive mechanisms behind updating belief in cancer misinformation. This will be conducted by exploring whether cancer-related misinformation is more difficult to correct than non- cancer related misinformation, and if so, why? This will be tested in both a general population and a population whose close relatives have cancer. In sum, this 5-year research and training plan will allow Dr. Swire- Thompson to establish an independent research program dedicated to understanding cancer misinformation on social media and its correction.

NIH Research Projects · FY 2024 · 2023-09

PROJECT SUMMARY Dr. Brady Post’s long-term goal is to explain how system-level integration affects quality and equity, specifically for rural residents. The American health care system has moved quickly toward greater integration between hospitals and physicians. As a way to organize healthcare services, hospital-physician integration has generated much controversy within the health policy community. Advocates hope for more efficiency and care coordination while skeptics raise concerns over costs and quality. This pressing policy issue, poised to be a defining issue in health care for decades to come, requires dedicated scholars. Dr. Brady Post has focused his scholarship on this topic since his doctoral studies at the University of Michigan and he continues to build the evidence base as an Assistant Professor at Northeastern University. A natural next step in his career is to advance the science on hospital-physician integration using a more sophisticated set of methodological approaches that will round out his skillset and, simultaneously, help to fill persistent gaps in the literature. Through a comprehensive training plan that includes formal coursework and mentored, project-based learning, this project will equip Dr. Post with (1) deep knowledge of rural health needs, policy, and service delivery (2) qualitative and mixed-methods techniques, and (3) survey design and measurement theory needed for survey research. This unique combination will position Dr. Post to direct innovative studies of how hospital-physician integration reshapes health care in rural communities. This training will build his skills as an independent researcher and prepare him to successfully compete for R01-level grants. This project will examine the effects of hospital-physician integration on care coordination, patient access to care, and disparities in health outcomes across rural and urban areas. It will leverage rich survey data to highlight patients’ perspectives and use in-depth interviews to highlight physicians’ and mangers’ perspectives. With this award, Dr. Post will bring new concepts and new approaches to the literature on hospital-physician integration at a time when the Centers for Medicare and Medicaid Services, the Federal Trade Commission, and other policymakers are actively weighing the benefits and drawbacks of increased integration. This proposal directly answers AHRQ’s request for health services research to advance health equity (NOT-HS-21-014) and focuses on the AHRQ priority population of rural residents. This award will support Dr. Post’s transition to independence, establishing him as a leading scholar on the quality and equity implications of increasingly integrated health care systems.

NIH Research Projects · FY 2025 · 2023-09

Project Summary Negative mood is a common feature of anxiety, depression, bipolar disorder, and schizophrenia, which inflict immeasurable human suffering along with a combined economic burden of $600 billion in the US each year. The brain basis of negative affect has been the focus of costly research efforts, but two critical barriers have slowed scientific discovery. First, there is no mechanistic explanation for how negative affect is caused in the brain. A solution to this barrier can be found in predictive processing, an emerging paradigm for unifying brain mechanisms across emotion, cognition, perception, movement, and other psychological domains. Predictive processing accounts posit that the brain continuously constructs prediction signals to control visceromotor and motor movements, while copies of these prediction signals anticipate incoming sensory signals from the body and the external world. Incoming sensory signals are thought to be relayed throughout the brain as prediction error signals. No study to date has examined negative affect in relation to the dynamics of signal flow within the specific architectural features of the brain. To surmount this barrier, I will take advantage of a conceptual innovation from our lab and thirty years of tract-tracing studies in mammals to test the hypothesis that prediction signals and prediction error signals can be traced across specific layers of cerebral cortex and subcortical structures. Briefly, prediction signals are thought to originate in deep layers of cortices that have less laminar development (e.g., anterior midcingulate cortex, aMCC, which is important for visceromotor control and affect) and arrive to subcortical structures (e.g., hypothalamus, involved in visceromotor control) and primary sensory cortices (e.g., primary visual cortex, V1). Interoceptive prediction error signals should originate from subcortical structures (e.g., hypothalamus) and other (exteroceptive) sensory prediction errors should originate in primary sensory cortices (e.g., V1), arriving to the upper and deep layers, respectively, of cortices with less laminar development (e.g., aMCC). In human subjects, these hypotheses remain to be tested due to a second barrier: neuroimaging methods have lacked sufficient spatial resolution to measure activity in deep vs. upper cortical layers and small subcortical structures. Newly developed ultra-high field (7 Tesla) fMRI techniques have sufficient resolution to overcome this barrier. With this methodological innovation, I will probe the mechanisms that cause negative affect in the circuitry outlined above via functional connectivity analyses of a 7T fMRI dataset our lab has curated. Ninety-two healthy subjects were instructed to anticipate visual or somatosensory stimuli (prediction period) that were either unpleasant or neutral and then were presented with the stimuli (prediction error period). In two specific aims, I will 1) measure dynamic prediction signals during negative affect, and 2) characterize prediction error signals during negative affect. The proposed research promises to deliver a new paradigm for studying the brain basis of negative affect, with the ultimate goal of developing targeted treatments for negative mood, a hallmark feature of many mental illnesses.

NIH Research Projects · FY 2025 · 2023-08

Abstract It is a formidable task to identify the molecular causes of complicated traits such as exceptional longevity (EL). The majority of machine learning algorithms generate mathematical correlations between genotypes and phenotypes, but may fail to infer physiologically significant causes. A mechanistic understanding of how individual molecular components work together in a system and how the system is affected and adapted to the molecular change requires knowledge of molecular interactions across all biological levels, from DNAs to RNAs to proteins to metabolites to organismal phenotypes. By integrating multi-omics data, recent approaches in multi-modal machine learning and multi-layer network model promise to address this deficiency. However, existing machine learning approaches are hampered by high-dimensionality, non-uniformity, numerous confounders, and biological differences in multi-omics data across data resources, data domains, and species as well as lack of interpretability due to the black-box nature of machine learning models. We will develop a transformative deep learning framework to address challenges for multi-omics data integration and predictive modeling of causal genotype-EL associations. This project is established on our substantial preliminary results, successes in systems pharmacology for Alzheimer's disease drug discovery and using C. elegans as disease and aging models, and close collaborations between experimental and computational laboratories. We shall overcome several obstacles in order to discover the molecular mechanisms of EL. We will develop and validate novel algorithms to 1) harmonize non-uniform data sets by removing environmental and biological confounding factors (e.g., age, species, etc.) and technical biases (e.g., batch effect), 2) explicitly model the biological information flow from DNAs to RNAs to proteins to metabolites to organismal phenotypes, and 3) determine causal genetic factors and molecular interactions underlying EL. Specifically, we will: (1) develop MuLGIT, a causal deep learning-powered cross-layer multi-omics harmonization and integration framework that follows the central dogma of biology for deciphering the molecular interplays underlying EL; (2) develop a transfer learning method PATH-AE for cross-species omics data integration and modeling for elucidating evolutionarily conserved and species-specific molecular determinants of EL; (3) identify molecular targets and pharmaceutical agents of EL by merging new methodologies for multi-omics data integration with state-of-the- art methods for chemical genomics and perturbation genomics; and (4) experimentally validate computational predictions using C. elegans models. Completion of this project will allow us to identify novel biomarkers, druggable targets, and pharmacological agents associated with remarkable lifespan (EL).

NIH Research Projects · FY 2025 · 2023-08

Racial disparities in end-of-life (EOL) care continue to persist in nursing home (NH) residents with Alzheimer’s disease and related dementia (ADRD). Recent studies reported that black NH residents with dementia were more likely to receive aggressive EOL care, including hospitalizations, ER visits, feeding tubes and aggressive medication therapy, compared to Whites. ADRD is a progressive, life-limiting syndrome without a curative treatment. Hence, hospice care is preferable for older adults with ADRD during EOL because it prioritizes comfort and quality of life by reducing pain and suffering. Medicare is the primary insurer of patients with ADRD and covers hospice for all with a 6-months or shorter prognosis. As about 90% of older Americans with ADRD are placed in NHs before death, it is critically important for Medicare policy makers to understand contemporary characteristics of racial disparities in hospice care in NHs in order to implement efficient policies to promote timely utilization of EOL hospice. Recently, there raised concerns about fraudulent/inappropriate patient selection practices that favored long-stay patients, particularly ADRD patients, because patients with longer hospice stays generated greater profits for hospices. To discourage hospice selection for long-stay patients, Medicare launched the 2014 Improving Medicare Post-Acute Care Transformation (IMPACT) Act to mandate auditing targeted hospices with high proportion of patient staying longer than 180 days. It remains unaddressed how racial disparities change with implementing IMPACT in NH residents with ADRD. The project’s overall goal is to improve EOL quality for NH residents with ADRD. The aims for this R01 proposal are to identify the effects of IMPACT on racial disparities in hospice care in NH residents with ADRD, and to characterize subgroups that are most likely to experience inadequate hospice care. The primary hypothesis is that racial disparities in EOL care increased persistently after IMPACT was implemented. We will employ mixed methods to accomplish the following aims (A) and hypotheses (H): A1. To examine impacts of IMPACT on racial disparities in hospice care in NH residents with ADRD. A2. To characterize phenotypes of NH residents with ADRD based on multimorbidity who have high risks for very short or very long hospice care and compare these phenotypes between Black and White residents. A3. To obtain perspectives of care providers in NHs about how hospice polices may influence hospice care referrals among residents with ADRD. An expert panel of clinical advisors will identify barriers to hospice care referrals, and potential mechanisms through which Medicare policies could reduce them. This study will evaluate Medicare policy on racial disparities in hospice care among a very vulnerable population: older adults with ADRD who reside in NHs. Results will also identify subgroups of these individuals at high risk of inadequate hospice stays. Findings will support clinicians with insight about how interventions to reduce racial disparities and improve EOL health quality can be targeted to individuals in high-risk groups.

NIH Research Projects · FY 2025 · 2023-08

Project Summary/Abstract Chromatin modifications that enhance DNA accessibility: Eukaryotic DNA is tightly packaged into nucleosomes, which form structural barriers to transcription, yet RNA polymerases effectively read through chromatinized DNA in cells. Crucial to this apparent paradox are ATP-independent histone chaperones, which include the heterodimeric FACT (FAcilitates Chromatin Transcription) complex and the monomeric proteins lens epithelium-derived growth factor (LEDGF) and hepatoma-derived growth factor 2 (HDGF2). Whereas FACT’s role as a histone chaperone is well-established, LEDGF and HDGF2 were only recently implicated as having histone chaperone activity. Furthermore, LEDGF and HDGF2 have also been implicated in modulating human immunodeficiency virus type 1 (HIV-1) DNA integration into chromatin, with LEDGF playing a dominant role, but the mechanisms remain unclear. We hypothesize that: (i) FACT facilitates chromatin remodeling by preferentially binding to nucleosomes destabilized by post-translational histone modifications, preserving histone-DNA interactions necessary for nucleosome reassembly; (ii) LEDGF and HDGF2 function as reader proteins, working selectively through their preferential binding to regions rich in H3K36me2/3 histone modifications, but have mechanistic similarities to FACT imparted by auxiliary domains; (iii) through their histone chaperone activity, LEDGF and HDGF2 proteins modulate HIV-1 DNA integration into actively transcribed genes characterized by chaperone-destabilized H3K36me2/3-rich chromatin. We propose two aims: Aim 1: Define the differential effects of FACT on the chromatin state To test the hypothesis that FACT acts as a chaperone by preferentially binding to unwound chromatin intermediates and chromatin destabilized by histone modifications, we will apply forces with optical tweezers to determine the equilibrium stability, fluctuational opening rate, and the ability of nucleosomes to reassemble after disruption in the presence of WT, mutant, and truncated FACT for histones modified through destabilizing acylation modifications. The results will reveal the extent to which FACT activity can be regulated by destabilizing histone modifications. Aim 2: Determine the mechanisms by which LEDGF and HDGF2 act as histone chaperones and facilitators of HIV-1 integration. To test the hypotheses that LEDGF and HDGF2 bind H3K36me2/3-containing nucleosomes and mediate nucleosome disassembly and reassembly and that LEDGF and HDGF2 direct HIV-1 DNA integration at H3K36me2/3-enriched loci through their nucleosome chaperone activity at these sites, we will measure the effects of wild type and mutant LEDGF and HDGF2 on nucleosome stability, dynamics and reassembly, as well as their effects on HIV-1 integrase binding. The results will determine the mechanism of LEDGF and HDGF2 nucleosome chaperone activity and the role played by histone methylation in regulating that activity.

NIH Research Projects · FY 2024 · 2023-07

Amyotrophic lateral sclerosis (ALS) is a fatal, rapidly progressing, neuromotor disease characterized by degeneration of upper (UMN) and lower motor neurons (LMN) affecting ~6/100,000 people in the US. ALS cases are projected to rise 69% globally (∼222K in 2015 to ∼376K in 2040). Most individuals with ALS die within 3-5 years of diagnosis, and as many as one third die within the first year. Yet, the time from onset of ALS symptoms to receive a diagnosis generally ranges from 8-15 months further crippling an individual’s quality of life, ability to plan, and treatment options in the already short time that remains. Finding a robust and reliable biomarker of corticospinal dysfunction in ALS is of paramount importance to improve diagnostic certainty at earlier stages, manage disease heterogeneity, track disease progression, improve patient stratification, and evaluate efficacy in clinical trials. Transcranial magnetic stimulation (TMS) is an effective tool to assess the functional integrity of UMN and LMN by measuring the degree of corticospinal drive – a measure called facilitation. Because the corticospinal system degenerates in ALS, the same tool can be used to quantify the degree of Failure of Facilitation as an indicator of the disease, paving the way for future studies to use this as an early biomarker of disease onset. In Aim 1 of the proposed project, we will test two key hypotheses in healthy individuals about the underlying relationship between facilitation in the corticospinal system and behavior. Then in Aim 2 of the proposed project, we will test whether individuals with ALS exhibit Failure of Facilitation, compared to those who are healthy or have an ALS mimic disorder, and whether the degree of Failure of Facilitation relates to the severity of the disease assessed with standard clinical batteries and neurophysiological tests of UMN dysfunction. This project will have significant scientific and clinical impact by advancing our understanding of the neural mechanisms for motor facilitation for voluntary force generation and providing an initial proof of concept that failure of facilitation is a useful biomarker of UMN dysfunction and disease progression in ALS.

NIH Research Projects · FY 2025 · 2023-07

PROJECT SUMMARY The survival of advanced or recurrent epithelial ovarian cancer (EOC) remains dismal due in part to the complex tumor–immune microenvironment (TIME). The presence of tumor-infiltrating T cells correlates with improved patient outcome in advanced EOC, yet checkpoint inhibitor immunotherapy has generally shown poor efficacy against EOC in clinical trials. A major challenge is the low number of T cells compared to EOC cells in the tumor that establish an immune-suppressive TIME. Tumor-targeted, cell-activatable photoimmunotherapy (taPIT, a near infrared phototherapy) may provide an alternative treatment approach that selectively destroys EOC cells expressing cell-surface epidermal growth factor receptor (EGFR) while enhancing the preservation of tumor infiltrating immune cells salient to an adaptive immune response (e.g., local cytotoxic T cells and dendritic cells). While photoimmunotherapy is not new, this is the first exploration of taPIT dose fractionation to selective eradicate EOC while stimulating immune cells at lower doses and preferentially sparing immune cells at higher doses. Based on our rich preliminary data, we propose that mathematical modeling-informed taPIT serves as a new paradigm for combined cytotoxic therapy and immunotherapy in metastatic EOC. Our overall goal is a novel experimental, simulation- and image-guided approach for utilizing the local selectivity of taPIT to prime “cold” TIME’s for immune checkpoint inhibition. This proposal contributes an innovative physical sciences approach to cancer therapy integrating mathematical modeling with a 3D culture model of the TIME and in vivo imaging experiments in immunocompetent mouse models of EOC, including in silico immuno-oncology modeling to optimize TIME composition-specific therapy; fractionated taPIT dosimetry to reduce the EOC cell burden relative to effector T cells and dendritic cells within the TIME; and, in vivo multiplexed fluorescence microendoscopy to interrogate the TIME of metastatic EOC within the peritoneal cavity. The unique ability to image micronodular disease will enable parameterizing a mathematical model with pre-treatment conditions and dynamic in vivo responses to therapy as a basis for the quantitative design of custom-tailored therapies. Our aims are (1) to train an in silico model by correlating clinical pre-treatment EOC TIME compositions with pathological anti-PD1 response; (2) to derive optimal fractionated taPIT protocols that shift “cold” to anti-PD1-sensitive “hot” TIME in vitro, and (3) to prime the TIME for anti-PD1 therapy in vivo. The concepts introduced here will ultimately enable taPIT–anti-PD1 therapy dosimetry that synergistically stimulates both local and distal immune-enhancing effects to impact disease sites missed by near infrared light, in combination with frontline surgical tumor debulking and systemic chemotherapy. The approaches developed here are translatable to other tumor sites that can be treated with taPIT, including head and neck cancers, skin cancers, and lung cancers.

NIH Research Projects · FY 2026 · 2023-06

Project Summary/Abstract Monocytes and macrophages function in diverse processes, from homeostatic maintenance to immune responses and tissue regeneration. These functions are coordinated with and strongly influenced by cellular metabolism via mechanisms that are increasingly studied and characterized in populations of macrophages. However, such studies mask the cell-to-cell variation which is an inherent property of macrophage diversity. Indeed, single-cell transcriptomics data have demonstrated that macrophage polarization is better described by continuous gradients rather than by discrete states amenable to isolation and population analysis. Yet, transcriptional measurements are insufficient to characterize the metabolic and protein networks that shape monocyte and macrophage diversity. To understand how these networks control macrophage polarization and functions, we propose to directly quantify proteins and regulatory signals (such as localization of key regulators, e.g., NF-κB) in primary human monocytes and macrophages responding to physiologically relevant metabolic environments. Furthermore, we will extend this single-cell analysis to the responses of these cells to pathogen-associated molecular patterns and damage-associated molecular patterns. These data will enable us to identify likely regulatory networks driving monocyte and macrophage responses to metabolic states and molecular patterns. Subsequently, we will test these networks via pharmacological and genetic perturbations. We are uniquely positioned to perform this research since we recently pioneered methods for quantifying thousands of proteins across many single cells. Furthermore, we have the required expertise in analyzing metabolic systems (including aerobic glycolysis, which is frequently associated with macrophage activation) and developing new algorithms for data analysis. This project will advance our understanding of macrophage immunometabolism and polarization, will introduce methods for more sensitive and accurate single-cell analysis, and will provide a proof-of-principle demonstration of the possibility to identify protein-mediated molecular mechanisms at single-cell resolution. We strongly believe that attaining these goals will have a transformative impact on biomedical research and will inform new and better therapeutic strategies.

NIH Research Projects · FY 2025 · 2023-06

Project Summary/Abstract While rural health care disparities are becoming a clear national crisis, the unique challenges faced by rural Americans are rarely discussed in today’s engineering classroom. Close to 20% of the United States population live in rural communities and face issues that are less common than their urban counterparts. According to the CDC, rural communities are at risk for a higher percentage of excess deaths from the 5 leading causes of death (cancer, heart disease, CLRD, unintentional injury, and stroke). Technical engineering innovations have the potential to address rural disparities. But to do so, explicit efforts must be made to expose future engineers to the healthcare challenges faced by rural America today. The goal of this proposal is to address the lack of focus on rural healthcare disparities in engineering classrooms. We believe that immersing bioengineering student groups in rural communities and hospitals will promote the identification of innovative engineering design solutions that can address rural healthcare disparities, encourage some students to work with the communities to find long term solutions, and establish a curriculum that supports career paths that seek to address rural healthcare disparities. The first aim of this program is to enhance rural health co-creation through partnerships. We will establish a clinical immersion co-creation program for a subset of NU bioengineering students in Maine rural health care facilities. Teams of students will work with Maine track medical students and clinicians over the course of 6 weeks to identify design opportunities emphasizing the concept of co-creation. The second aim is to scale the capstone design program at Northeastern by establishing a team-based approach to integrating the broader NU bioengineering community into design for rural health. Recognizing that not all students can participate in the clinical immersion, the rural healthcare teams will create a compendium of rural healthcare needs to serve as design projects for the capstone bioengineering course, allowing up to 36 students per year to participate in rural engineering design. The final aim is to Establish a pathway for students to implement their solutions. We seek to avoid a “hit and run” problem, where students enter a community they seek to serve only to leave and not sustain the relationship and implement sustainable solutions. After completing capstone design, students will have the opportunity to enter a one-year Roux Institute startup residency program that connects early- stage businesses to resources and best practices from across the university’s global network, providing a platform for implementing solutions in Maine. We believe that the proposed approach will establish a model for not only distributing students throughout the Maine rural health care system for opportunity identification but the development of the relevant curriculum to support students in co-creation and innovative design for rural communities.

NIH Research Projects · FY 2025 · 2023-06

Project Summary The cornea and sclera are the principal load bearing members in the tough fibrous ocular tunic which we consider to be an integrated mechano-biological structure. During development, the shape of the eye has been tuned to conform to a specific mechanical environment through a long time-scale integration of its loading history with its initial genetic patterning. Although mechanics are known to contribute to the development of many connective tissues, the ocular globe is particularly sensitive to pressure (tensile wall stress) during the expansive phase of growth. Even in the mature ocular tunic, mechanical instabilities often manifest as conditions which disrupt vision (e.g. myopia, keratoconus, post-LASIK ectasia, tractional retinal detachments and glaucoma). While the underlying causes of tissue structural instabilities are poorly understood, we suspect that they are mechanobiological in nature and potentially reflect an imbalance in the tensional homeostasis that exists between mesenchymal cells, their local ECM and the global mechanical environment. We know that during development, disruption of the mechanical connection between fibroblastic cells and their ECM severely retards ocular growth in a manner analogous to pressure loss, suggesting that mechanical communication is critical to proper ocular morphogenesis. However, the effect of mechanical forces on the mechanisms which drive tissue formation and growth are not well characterized. It is remarkable that we still do not fully understand how the most important structural molecule in vertebrates, collagen, is efficiently assembled into highly-organized, functional, load-bearing tissues which are massively expanded into macroscale structures during growth. However, if we are able to uncover new mechanisms which control tissue formation and growth, we will have access to information which can inform therapies for a variety of pathological conditions including fibrosis, myopia, keratoconus and potentially, glaucoma. Additionally, if we understand how tissue is produced, then there will be implications for engineering corneas de novo and for improving approaches to regenerative corneal medicine. In the proposed work, we plan combine our human cell culture model of corneal stromal tissue elaboration with live-cell mechanodynamics imaging to directly observe single collagen molecules during their transition from solution to fibrils. We will thus directly test a new hypothesis which directly couples local and globally applied forces directly to molecular assembly of collagen during fibrillogenesis and growth. The working hypothesis for this proposal is that force causes corneal stromal ECM elaboration to regulate fibril assembly, remodeling and growth. If the hypothesis is correct, there are myriad mechanotherapeutic opportunities and more critically, our basic understanding of collagenous tissue formation and growth, will be fundamentally altered.

NIH Research Projects · FY 2026 · 2023-05

PROJECT SUMMARY New generation pod-style nicotine salt e-cigarettes (e-cigspod) are popular because of their sleek design and ability to deliver nicotine levels similar to those of conventional tobacco cigarettes (cigs). Nicotine salts are easier to inhale than the free-base form of nicotine found in previous generation e-cigs. While the long-term health impact of cig smoke is well known, the consequences of chronic e-cig use remain in question, and data is needed to justify immediate FDA regulations. A substantial portion of smokers are unsuccessful in using e-cigs to support their cessation efforts and instead become dual users. The chemical profile of cig smoke and e-cigpod aerosols is different, which suggests that the health effects of chronic smoking and vaping may not fully overlap. Building upon this, we hypothesize that cig smoking and e-cigpod vaping, are independent risk factors for cardiopulmonary disease, whose superposition exacerbates the maladaptive remodeling of the lungs, heart, and vasculature compared to either practice alone. To test this hypothesis, we will expose hypercholesterolemic and wildtype mice to nebulized nicotine, aerosolized solvent carrier, e-cigpod aerosols, and cig smoke (naïve mice), or a combination of the two (both naïve and previously cig smoke-exposed mice) and we will compare the structural and functional remodeling of the cardiovascular and respiratory systems. We will generate e-cigpod aerosols from pods in Tobacco flavor at 5% nicotine strength. Motivated by the idea that smokers who use e-cigs as cessation aids may vape until they satisfy their nicotine cravings, we will perform experiments to achieve equal cotinine bioavailability in the mouse blood, while maintaining the same daily duration of exposure. We will measure the mechanical properties of the of aorta (tissue stiffness, distensibility, and elastic storage), heart (fractional shortening and ejection fractions), and lungs (resistance and elastance). We will characterize tissue microstructure (air space sizes, collagen content, and elastic fiber integrity) to highlight the factors that most contribute to the observed functional changes. Knowledge gained from this project will provide scientific evidence in support of data-driven e-cig regulation under The Family Smoking Prevention and Tobacco Control Act (FSPTCA), specifically concerning the health risks of dual combustible and electronic cigarette use.

NIH Research Projects · FY 2025 · 2023-05

PROJECT SUMMARY: Effective gait rehabilitation must be engaging, intense, repetitive, and focused on purposeful movement. However, patients after stroke can have neuromuscular or cognitive impairments that make this ideal unattainable without physical assistance. Paradoxically, this assistance can hinder motor recovery when patients habituate and reduce their sensorimotor engagement (the active correction of motor errors). To counter habituation and improve learning during assisted gait training, this project newly combines two theoretical frameworks: one from social psychology governing how social pressure impacts motivation, and another from neuroscience describing motor learning as processes of error-driven adaptation and reinforcement learning. Positive social pressure from working in a team can boost motivation compared to working individually, which may improve rehabilitation by enhancing active error correction and increasing the value of successful actions. This motivation enhancement in a team setting is termed the Köhler effect and is moderated by two mechanisms: upward social comparison (comparing self-performance to others) and indispensability (when success requires effort from all team members). Although these mechanisms have robust positive effects in exercise and cognitive tasks, their role in motor learning is unknown. Addressing this gap could improve rehabilitation care for millions who annually acquire and live with impairments from neurological injury. This fellowship’s complementary specific aims target the impact of each Köhler mechanism on locomotor rehabilitation. Aim 1 investigates how promoting upward social comparison affects sensorimotor engagement and motor learning during assisted locomotion and Aim 2 investigates the effects of indispensability. To enhance scientific rigor, a new model system is used: healthy participants walk on a treadmill and receive a standardized, temporary neuromotor impairment by electrically activating their right hamstrings during leg swing. This perturbation decreases participants’ net mechanical work during swing and shortens step length. The goal is to reach a target mechanical work while viewing step-by-step work feedback. Unlike a kinematic target, which a therapist can correct, the work goal can only be achieved if the participant contributes a significant amount of effort to the rehabilitation task. Showing the separated participant and therapist work contributions is expected to evoke upward social comparison. In Aim 2, therapist assistance is provided through a telerobotic system to rebalance effort contributions, making the therapist “weaker” and participants indispensable. This project is accompanied by a robust fellowship plan of interdisciplinary training that bridges neuroscience, rehabilitation, social psychology, engineering, and post-acute stroke care. The major theoretical advancement will be a new understanding of how sensorimotor engagement and motor learning are influenced by positive social pressure. The results will serve to increase the effectiveness of rehabilitation interventions by shifting patient perceptions, so they view themselves not as lesser contributors to a shared rehabilitation goal but as indispensable to success.

NIH Research Projects · FY 2026 · 2023-04

PROJECT SUMMARY/ABSTRACT Mammalian cells expend large amounts of energy into generating enzyme-mediated RNA chemical modifications that can change the base-pairing, RNA structure, or recruitment of RNA-binding proteins, among other elusive roles. Pseudouridine (ψ)-modified mRNAs are more thermodynamically stable, more resistant to RNAse-mediated degradation, and have the potential to modulate immunogenicity and enhance translation in vivo. However, ψ detection is extremely challenging: ψ modifications do not affect Watson-Crick base pairing and are indistinguishable from uridine when using hybridization-based methods. Further, since ψ is an isomer of uridine, detection using mass spectrometry requires non-quantitative chemical derivatization methods. While recent studies have shown that RNA modifications can be detected through direct RNA nanopore sequencing by monitoring basecalling errors, we have recently shown that the accuracy and fidelity of this approach is relatively low and sequence dependent. Our team has recently used a ligation approach to produce synthetic mRNA controls that contain single ψ sites within relevant transcripts mammalian cells. Using these synthetic controls we performed nanopore-based RNA sequencing and developed computational tools that increase the accuracy of ψ-calling to 90+%, depending on the specific sequence. We are basing our work on our recent finding that achieving ψ quantification requires sequence-specific training using unique signal parameters. The initial success of our team has laid the foundation to 1) generate an expanded set of barcoded synthetic RNA constructs that contain single ψ sites, 2) obtain a rigorous set of quadruplicate nanopore runs with ~50,000 single-molecule reads per construct, 3) develop computational tools to allow highly accurate sequence-specific ψ-calling. We will develop a gold-standard set of synthetic mRNA transcripts as a training molecular set for quantitative ψ profiling in direct RNA nanopore sequencing of human transcriptomes. The molecular set will allow quantitative profiling of hundreds of putative ψ sites across mammalian samples. This proposal will serve an unmet need by addressing a critical bottleneck: the lack of available modified RNA modification gold standards, i.e., RNA molecules that contain a site-specific and structure-specific modification. In this collaborative project we will develop a complete pipeline for synthesis of gold standard molecules; use these molecules to measure the nanopore signals that ψ modifications produce; develop a machine-learning tool to accurately quantify these modifications; profile site-specific ψ modifications in various cell lines to obtain ψ-maps that can be used to assess relationships of ψ modifications with phenotypes.

NIH Research Projects · FY 2024 · 2023-03

PROJECT SUMMARY Adolescent major depressive disorder (MDD) is common and debilitating. Presently, gold-standard treatments are only effective for approximately half of patients, underscoring the need to develop novel interventions, particularly to target core underlying mechanisms and more effectively treat this recurrent disorder. Rumination, the tendency to perseverate about depressive symptoms, contributes to MDD onset and predicts treatment non-response and relapse. At the neural level, rumination is characterized by elevated functional connectivity within the default mode network (DMN), and prior research also has consistently demonstrated patterns of DMN hyperconnectivity in MDD. Interestingly, mindfulness meditation, which trains attentional focus to the present moment, reduces perseverative thinking, ruminative tendencies, and depression symptoms. Further, our research and others have shown that adolescents can apply mindfulness practices to decrease perceived stress, increase sustained attention, and suppress DMN activity. Although mindfulness has profound mental health benefits, for some, mindfulness alone may not be sufficient to mitigate ruminative tendencies during a depressive episode. That is, MDD symptoms, including reduced motivation, inattention, and lack of self-efficacy, may impede a patient’s progress in successfully acquiring and utilizing mindfulness strategies necessary to change perceptions about one’s environment and relationships. To directly address this challenge, we propose using real-time fMRI neurofeedback to enhance the acquisition and utilization of mindfulness skills to better target DMN hyperconnectivity, rumination, and depressive symptoms. We developed a novel, 15-minute mindfulness-based, real-time neurofeedback (mbNF) paradigm whereby people observe a visual display of their brain activity and practice mindfulness to volitionally reduce DMN activation. In the R61 phase, 90 adolescents (ages 13-18) diagnosed with MDD will complete a 45-minute mindfulness training outside the scanner. To test target engagement of reducing DMN hyperconnectivity and optimal dosing, adolescents will then be randomized to receive either a 15- or a 30-minute mbNF session (n=45/dose group). If we meet our Go criterion (i.e., significantly reducing DMN hyperconnectivity), we will then proceed to the R33 phase to investigate whether mindfulness with mbNF outperforms mindfulness only. Thus, a new sample of 120 depressed adolescents (ages 13-18) will participate in a double-blind randomized clinical trial and receive mindfulness with the optimal mbNF dose (i.e., per the R61) or mindfulness only (n=60/group). We will test–using clinician-administered instruments, self-reports, and ecological momentary assessment– whether compared to mindfulness only, mindfulness with mbNF contributes to a greater reduction in clinician assessed depression symptoms (primary outcome) as well as decreased rumination (secondary outcome) across the post-treatment, 1-month, and 3-month assessments. As a whole, mbNF is directly in line with precision medicine initiatives, and if successful, could revolutionize clinical care for depressed adolescents.

NIH Research Projects · FY 2026 · 2023-03

PROJECT SUMMARY Dravet Syndrome (DS) is a devastating form of epilepsy caused by loss of function of NaV1.1 (80-90% of cases), the predominant voltage-gated Na+ channel expressed in inhibitory (GABAergic) interneurons in the hippocampus and prefrontal cortex. This causes a decrease in the release of inhibitory neurotransmitter (GABA), resulting in hyperexcitability. The disease manifests itself within the first year of life and is usually triggered by hyperthermia causing frequent and prolonged seizures that result in a host of health problems including developmental delay, speech impairment, ataxia, hypotonia and sleep disturbances. Two small molecule monotherapies have been approved recently by the FDA: Epidiolex (Cannabidiol or CBD) in 2018 and Fintepla (fenfluramine or FA) in 2020 for patients two years of age and older. Even though they reduce the frequency of seizures, these drugs at their effective dosages cause multiple side effects. Their mechanism(s) of action to reduce epileptiform activity remain(s) unknown. G protein-gated inwardly rectifying K+ (GIRK) channels have been strongly implicated in epilepsy. They are activated by the Gβγ dimer of G proteins and by [Na+]i in a synergistic manner. The basis of synergism lies in that they each work allosterically to control predominantly the two channel gates: Gβγ, the membrane gate and Na+, the cytosolic gate. In the case of the GABAergic interneurons, we hypothesize that NaV1.1 and GIRK1/2 are coupled, such that the Na+ entering through NaV1.1 promotes GIRK1/2 activity to hyperpolarize the cell and ensure removal of NaV1.1 inactivation for fast spiking. In DS this mechanism becomes compromised causing cell depolarization and inactivation of voltage-gated channels at large present in GABAergic neurons failing to compensate for the loss of NaV1.1. We use GAT1508, a specific activator of GIRK1/2 to compensate for the compromised Na+ entry. Since GAT1508 opens the cytosolic gate, we ask whether it synergizes with CBD (via CB1R) and FA (via 5-HT1DR) to open more fully the membrane gates. In Aim 1, experiments designed to test the hypothesis are aimed at the cellular level in both heterologous expression and in native GABAergic neurons. In Aims 2 and 3, we utilize a DS mouse model, heterozygous for the Scn1a gene that encodes NaV1.1, and in Aim 2, we test the hypothesis at the brain slice level, where synapses and transmitter release remain intact, and compare the DS model to a wild-type animal model. In Aim 3, we pursue experiments at the whole animal level (DS model), using simultaneous EEG and 2-photon microscopy to monitor the neuronal circuits involved.

NIH Research Projects · FY 2025 · 2022-09

Project Summary / Abstract A central problem in biology is to understand how genomic variation affects genome function to influence phenotypes. Key challenges and opportunities lie in linking genomic variants to phenotypes, human health, and disease. Because it is not feasible to experimentally probe all genomic variants of interest in all contexts, improved computational methods to accurately predict the impact of unknown genomic variants are necessary. The aim of this research proposal is to gain mechanistic understanding of functional genomic interactions and ultimately to develop computational approaches to model and predict relationships among variation, functional elements, genome function, and phenotype. Two recently acquired key assets will be used to infer distal functional interactions among DNA elements: i) 3D genomics data and ii) multiple genome alignments. High- resolution contact mapping experiments (Hi-C and similar methods) have shown that the structural ensembles of chromosomes are fluid and yet specific to cell type and phase of life1. These ensembles of partially organized structures bring sections of DNA separated by great genomic distance into close spatial proximity and play an important role in controlling gene transcription 2,3. By measuring the frequency of physical contacts among DNA elements, DNA-DNA proximity ligation assays offer insight into the existence of functional interactions among the same elements, even when the nature of the interaction is unknown. In the last few years, there has been an explosion of activity directed toward assembling the genomes of many species 35–37. Hundreds of newly assembled end-to-end genomes constitute a dataset of transformative importance in studying the general operating principles of genomes across the tree of life using evolutionary information. This proposal aims to combine data from proximity ligation assays and coevolutionary information extracted from multiple genome alignments to infer the network of functional interactions among DNA elements. The computational approach will be based on Direct Coupling Analysis 29–32 (DCA) and other machine learning methods. The PI has previously employed DCA to study genome architecture 33 as well as in other contexts 34, and has already made important contributions to the field of 3D genomics.

NIH Research Projects · FY 2025 · 2022-09

PROJECT SUMMARY (See instructions): Autism Spectrum Disorder (ASD) is one of the most common childhood disorders (1 in 44 ). Individuals with ASD have a higher prevalence of challenging behavior (e.g., aggression, self-injury, emotion dysregulation) that interferes with adaptive development, ranks among the most common causes for referral to behavioral healthcare services, and incurs high healthcare costs. Over the past four years, the team at Northeastern University has made significant research progress developing machine learning procedures that automate the detection of challenging behavior onsets in individuals with ASD using wearable biosensor data (cardiovascular, electrodermal, and physical activity). Despite our promising results, issues still need to be addressed to enable practical daily use in real-world contexts. This includes the need for extensively labeled data, individual calibration from population models to specific individuals, and handling the non-stationary nature of challenging behavior and physiological data. This project aims to advance fundamental machine learning theory and techniques that facilitate rapid model individualization and continuous online model adaptation with little or no labeled data. To this end, we will contribute to areas including domain adaptation, transfer learning, lifelong learning, and robust modeling and inference. Three Specific Aims guide the project: (1) Rapid physiological and behavioral data model individualization; (2) Continuous lifelong physiological and behavioral data model adaptation; and (3) Validation of model individualization and adaptation techniques with prospective data collected in a clinical setting from our partners at the Marcus Autism Center at Emory University who specialize in functional analysis of challenging behavior in individuals with ASD. Across these Aims, we will explore applications of semi-supervised learning theory, data importance weighting, Support Vector Machines, neural network models, Hierarchical Markov-Modulated Point Process Models, and Bayesian evidence fusion. The modeling and inference techniques we develop will be of general applicability to other health application contexts involving event prediction (e.g., seizure detection) and human action/decision-making (e.g., intensive care unit triage).