Carnegie-Mellon University

NIH Research Projects · FY 2025 · 2022-07

PROJECT SUMMARY Increasing the efficiency of computational methods has been instrumental to extracting insight from genomic data. Fast aligners such as MUMMER, fast k-mer counters such as JELLYFISH, fast expression quantifiers such as SAILFISH and SALMON, and high-quality efficient genome assemblers such as MASURCA have been crucial to unlocking the potential of genomic and metagenomic data. Nevertheless, computation remains a time and cost bottleneck in many application areas. Algorithmic sketching methods, such as the minimizer schemes, have been a useful technique for achieving improved computational efficiency. However, despite their importance, these sketching techniques are understudied from a theoretical perspective and underused from a practical perspective. We propose to design, implement, test, and validate new sketching approaches based on significant extensions to the successful minimizers sketching schemes, greatly increasing the flexibility of these approaches and ex- panding their use into new areas including handling high-variance or highly repetitive sequences, and providing a new, standard sketching toolkit for genomic method designers and software implementors. These extensions, collectively referred to as marker selection schemes, will enable faster alignment, clustering, and assembly of genomic sequences, and will spur further computational innovation in genomic applications. To inform and validate this algorithmic work, we propose to enhance three important and broad areas of genomic computational methods. First, we will extend the widely-used MUMMER aligner with a number of application- specific “modes” that exploit these new and existing sketching schemes to achieve enhanced efficiency and greater sensitivity. This will ensure continued development and enhancement for additional applications of this important computational tool. Second, we will enhance the MASURCA genome assembler with updated in- tegration with the new MUMMER. Third, we will use the developed marker selection schemes and additional algorithmic ideas based on geometric embedding of sequences to develop more accurate, fast estimators of distances between genomic sequences. These approximate distance estimators are essential for a number of metagenomic applications including species classification, clustering, and search. We will advance the compu- tational accuracy of these tasks through these improved estimators. This project will result in a deeper toolbox of genomic sketching and distance estimation algorithms, software libraries encoding these new algorithms for wider use by the community, and an improved suite of genomic software, including enhancements to a widely used aligner and assembler and improved accuracy in existing and new metagenomic software.

NIH Research Projects · FY 2025 · 2022-07

Project Summary Rationale: This proposal seeks NIH support for a T32 training program in Big Data Systems Neuroscience, where students with strong backgrounds in neurobiology will receive training in quantitative methods and experimental design. Rapid technological advances in experimental neuroscience have enabled the collection of increasingly large neural data sets, but students in traditional neuroscience programs rarely receive formal training in the analytical approaches needed to understand such complex data sets. This knowledge gap slows progress and imperils the rigor of scientific research. Bridging this knowledge gap requires collaboration between data scientists and experimentalists, which is at the heart of this T32 training program. Carnegie Mellon University (CMU) is a world leader and innovator in quantitative fields such as machine learning, computer science, and artificial intelligence, and recently, neuroscience has emerged as a field for strategic growth at the university. The training program will be housed in CMU’s Neuroscience Institute, established in 2018 to promote interdisciplinary research and training in big data neuroscience. Goals: As one of the first programs of its kind, the T32 training program in Big Data Systems neuroscience will create a unique collaborative and multi-disciplinary training experience with three overarching goals, to: 1) provide Ph.D. students interdisciplinary, cutting-edge training in experimental and quantitative aspects of systems neuroscience research and its responsible and ethical conduct 2) provide Ph.D. students with thoughtful, career-specific mentoring and additional professional development opportunities to increase their retention in systems neuroscience, as well as more broadly in the biomedical workforce 3) Provide four of the most accomplished students in later years of the program with a 24-month fellowship to develop research projects integrating theoretical and empirical approaches 4) enhance interactions, collaborations and the overall quality of systems neuroscience research among trainees and faculty at CMU and the University of Pittsburgh.

NIH Research Projects · FY 2025 · 2022-06

PROJECT SUMMARY/ABSTRACT Our level of arousal influences almost all of the cognitive and behavioral abilities that allow us to engage with our environment. A variety of terminology has been used to describe arousal-related changes in cognition, including wakefulness, internal state, engagement, and alertness. Despite this ambiguity, it is clear that arousal is a global phenomenon that affects brain activity, systemic physiology, and behavior. Arousal deficits in behavior and systemic markers, such as heart rate and pupil diameter, are present in many neurologic disorders, including Alzheimer’s, schizophrenia, and attention deficit disorder. In order to use these non-invasive systemic measures of arousal as diagnostic markers, we need to understand how arousal level influences brain activity in neurotypical cognition and in turn leads to changes in behavior. The goal of this proposal is to investigate changes in brain activity that occur with fluctuations in arousal level. Specifically, we will study slow changes in simultaneously measured cortical neural population activity and cerebral hemodynamics. We will track arousal level using pupil diameter and heart rate. Our overarching hypothesis is that we can link minute-to-minute arousal fluctuations with slow changes in neural population activity and cerebral hemodynamics. In the first aim, we will study internally driven variations in arousal level that occur over hours and across the circadian cycle. In the second aim, we will probe arousal level using an external stimulus to assess whether we can modulate arousal- linked drifts in neural population activity and hemodynamics. Our strategy to record simultaneous population activity and hemodynamic changes in context of systemic parameters and behavior will allow us to comprehensively probe arousal-related changes and integrate arousal effects across the brain, body, and behavior. Results from this proposal will help to understand how deviations in arousal, measured via behavior and systemic physiology, could serve as biomarkers of changes in global brain activity that are present in neurologic disorders. This proposal will be completed in the interdisciplinary and collaborative training environment of Carnegie Mellon University and its Center for the Neural Basis of Cognition that consists of a large community of neuroscientists and engineers investigating brain activity underlying cognition and brain disorders. Under the guidance of Dr. Matt Smith, an expert in systems neuroscience, and Dr. Jana Kainerstorfer, an expert in biomedical optics, this proposal will provide me with a multimodal skill set in electrophysiology, optics, and psychophysics to probe cognitive function. These skills will help me progress towards my goal of becoming a physician scientist studying the interaction between the brain, systemic physiology, and behavior to identify biomarkers of cognitive disorders.

NIH Research Projects · FY 2026 · 2022-05

While the regulation of gene expression is well recognized as being important to human disease, there is a fundamental gap in understanding the role of post-transcriptional processes. Because post-transcriptional processes are critical for the regulation of protein production, addressing this knowledge gap will facilitate better models of the relationship between genotype and phenotype. Like transcription, mRNA translation is regulated by cis-acting sequences and trans-acting factors. upstream Open Reading Frames (uORFs) are cis-acting regulatory element found in most human genes, and some disease-linked mutations appear to alter the presence of uORFs. The primary focus of my laboratory is to determine how cis-acting sequences and trans-acting factors control translation. Our goal for the next funding period is to determine the functions of human uORFs and evaluate their regulation by trans-acting RNA binding proteins (RBPs). To accomplish this, we have adapted our yeast Massively Parallel Reporter Assays (MPRAs) for use in human tissue culture. In addition, our computational analysis has identified thousands of primate-conserved human uORFs, many of which have conserved RBP binding sites downstream of their start codons. The vast majority of these uORFs have not been functionally studied. Our innovative approach combines exquisite systems biology tools with cutting-edge computational modeling to investigate the functions of these ubiquitous cis-regulatory elements. The proposed research is significant because it is expected to fundamentally advance our understanding of human gene regulation.

NIH Research Projects · FY 2026 · 2022-03

Project Summary/Abstract The past decade has seen outstanding advances in the genetics of autism spectrum disorder (ASD). Most of this progress has occurred by the study of rare genetic variation, especially de novo variation, with the Autism Sequencing Consortium (ASC) playing a central role. The ASC represents a coordinated international effort to identify ASD risk genes. In our most recent, unpublished, analyses of 72,410 individuals from ASD families, we identified 185 genes associated with risk (FDR < 0.05). Some of these genes have been linked to a broad array of developmental disorders, while others have not. Based on these results, we posit that some risk genes alter the core features of ASD, while creating fewer perturbations to other features of development: discovery of such genes will provide deeper insights into pathways disrupted in ASD. We will build on this progress by analysis of sequence data from three resources: ASD subjects and families; subjects with other developmental and neuro- psychiatric disorders; and subjects from population samples. We plan new research focusing on interpretation of rare variation, including single nucleotide variation (SNV), indels, and copy number variation (CNV). Our key targets are inherited variants, including X-linked inherited variants, which to date have shown very little signal, and missense variants, for which signal has been confined to highly conserved substitutions. We anticipate doubling the number of ASD genes discovered, ~ 400, by increasing the number of families analyzed and by refined methods to interpret inherited and missense variation. And, in parallel, we expect to resolve critical as- pects of ASD genetic architecture and to unveil key aspects of what makes ASD and its core features – social deficits and restrictive and repetitive behaviors – different from other neurodevelopmental disorders. To discover ASD risk genes with a distinct effect on ASD, we have the following specific aims: 1) To amalgamate existing and emerging whole exome and whole genome sequence data; 2) To develop new analytical methods and analyze the accumulated sequence data; and, 3) To contrast ASD and other neurodevelopmental disorder risk genes, examining developmental profiles, cell types implicated, and whether variants in the same gene differ in how they affect risk for ASD and other neurodevelopmental and psychiatric disorders. With this new research we will accelerate our overall objective, which is the identification of ASD genes, thereby facilitating our long- term goal of building the foundation from which therapeutic targets for ASD emerge. Our rationale is that the identification of genes conferring significant risk to ASD and associated neurodevelopmental disorders can form the basis of studies to understand pathogenesis, as well as the basis for novel therapies. Our central hypothesis – formulated based on results over the past decade – is that rare and common variation contributes additively to risk for ASD, but only certain rare variants confer substantial risk. The research proposed is innovative, in our opinion, because it uses groundbreaking and novel statistical methods for identifying risk variants for ASD.

NIH Research Projects · FY 2026 · 2022-02

ABSTRACT Synaptic plasticity in neocortical neurons is intimately tied to learning and memory. Decades of research in acute brain slices have characterized the patterns of spike timing required to evoke synaptic change in minute detail, but it remains unknown whether these conditions occur and are sufficient to drive synaptic plasticity in the living brain. Indeed, in vivo recordings indicate that neocortical neurons live in an environment of profound inhibition that lowers overall firing rates and prevents plasticity. How then do cortical neurons escape this inhibition to encounter appropriate conditions for plasticity during learning? New evidence suggests that parvalbumin (PV) GABAergic neurons may play a dominant role in regulating cortical activity and controlling network rewiring, particularly at the early stages of learning. Using a multiwhisker stimulus coupled to a water reward, we have developed a paradigm for sensory association learning that drives rapid changes in excitatory synaptic strength in mouse barrel cortex. Importantly, our new data indicate that PV output to neocortical pyramidal neurons is markedly suppressed at the earliest stages of sensory training. Our experiments will integrate in vivo and acute brain slice recordings to test the hypothesis that PV neurons are a dominant regulator of sensory- evoked activity in mouse barrel cortex. We propose that reward-related acetylcholine release indirectly suppresses PV neural firing to depress PV output and increase sensory-evoked activity during learning. Our experiments will identify mechanisms for cortical disinhibition that facilitate experience-dependent synaptic plasticity in sensory cortex.

NIH Research Projects · FY 2025 · 2022-02

PROJECT SUMMARY The three-dimensional (3D) genome organization in the nucleus is of vital importance to genome function. The vast majority of the existing 3D genome studies, however, are based on population-based assays that are unable to unveil the functional roles of 3D genome structures at single-cell resolution in complex tissues. Recent advent in single-cell Hi-C (scHi-C) technologies has enabled genomic mapping of chromatin interactions in individual cells, but the analysis of scHi-C data remains a significant challenge. In particular, computational methods that can effectively analyze scHi-C data to extract multiscale 3D genome features are significantly lacking, limiting our ability to reveal the variability of structure and function connections in heterogeneous cell populations. The overall objective of this proposal is to develop state-of-the-art computational tools for scHi-C data analysis that effectively identify multiscale single-cell 3D genome features and connect them to genome function. Specifically, we will (1) develop algorithms for scHi-C data processing and imputation to delineate multiscale 3D genome features; (2) develop computational methods to connect 3D genome structure and function in heterogeneous cell population; and (3) develop an integrative visualization platform to navigate single-cell 3D genome organization. The methods developed in this project can be applied to all types of scHi-C data generated by different single-cell chromatin interaction assays to reveal 3D genome features at multiple scales, quantifying their variability and predicting their functional outcomes. The new tools and resources from this project will be publicly accessible through our new visualization platform that provides integrative and interactive navigation of scHi-C data and other data types. Overall, our project will greatly facilitate the use of scHi-C data by the broad scientific community and be of high value to a diverse group of biomedical researchers.

NIH Research Projects · FY 2026 · 2021-12

PROJECT SUMMARY This project will test the hypothesis that DNA damage in cardiomyocytes activates p53 leading to mitochondrial alterations and secretion of paracrine factors that drive heart failure. The premise for this has been established from our preliminary data and from the work of others. First, DNA damage and activated DNA damage response (DDR) have been observed in cardiovascular disease (CVD) in humans. Second, studies also show evidence that multiple cell types in the cardiac unit, including cardiomyocytes (CM) and cardiac fibroblasts (CF) display markers of DNA damage and cellular senescence in several disease pathologies. Third, we have recently identified that nuclear DNA damage drives dilated cardiomyopathy. Specifically, cardiomyocyte-depletion of the DNA repair endonuclease, ERCC1-XPF in mice, upregulates the DNA damage response gene, p53, and leads to irregular mitochondrial cristae, accumulation of lipids and increased oxidative stress. Additionally, there is an increase in several cardiac failure and senescence associated markers. However, the exact molecular underpinnings and cell-specificity of these DNA damage-induced changes is poorly understood. One barrier to addressing this question in vivo has been lack of appropriate tools, where DNA damage can be introduced in only one cell type (e.g., CM) and its effect on CF and cardiac function can be investigated. Additionally, 2D cell culture and co-culture systems fall short, as they cannot reproduce tissue dynamics present in a cardiac unit. Herein, we have developed several tools enable the study of cell-cell communication of 3D multicellular system. Specific Aim 1 will map the molecular, functional, and architectural changes upon loss of ERCC1 in CM. In this aim, we will test the mechanistic role of p53 and reactive oxygen species on a number of cellular and mitochondrial parameters, as well as cardiomyocyte electrophysiology. Specific Aim 2 will test whether stochastic, spontaneous DNA damage in the CM or CF drives cardiac electromechanical dysfunction in a cell- autonomous or cell non-autonomous manner through a paracrine effect on neighboring cells. Here, we will analyze the pathological secretome upon genotoxic stress, as well as test the role of eliminating senescent cells on cardiac health. This work is technically innovative as it uses a number of unique tools including concomitant optical and bioelectrical measurements in 3D cardiac organoids. These contributions will be significant because DNA damage is unavoidable and intimately linked to cardiac health and disease. Our team is uniquely qualified to perform this work, with expertise in DNA damage/ repair, cellular senescence, nanofabrication, human iPSC- derived cardiac tissue engineering, and data science. This analysis, we believe, will increase our fundamental understanding of the connection between DNA damage and heart disease and potentially pave the way for new treatment strategies.

NIH Research Projects · FY 2025 · 2021-09

Non-invasive neuromodulation approaches have been developed to enable the modulation of neural tissue without necessitating invasive surgical procedures. Low-intensity transcranial focused ultrasound (tFUS) neuromodulation has proven its efficacy and precision in modulating the brain, from the neuron to circuit level. However, there is an urgent unmet need to elucidate the in vivo neuronal and inter-neuronal effects of the tFUS neuromodulation, thus advancing the translational application of tFUS neuromodulation on humans. We propose to investigate the in vivo neuronal cell-type specific response and long-term plasticity effects of tFUS by systematically examining tFUS parameters in both anesthetized and awake rat models using a novel, cutting-edge 128-element random ultrasound array for rodents. The proposed experimental investigations are built upon our preliminary explorations and rigorous understanding of how different low-intensity tFUS parameters lead to unequal responses among unique in vivo neuron populations and the sustained alteration of synaptic connectivity in anesthetized rodent models using intracranial recordings. We will address the following specific aims. Aim 1. Characterization of intrinsic in vivo cell-type specific response of somatosensory cortical circuits to tFUS stimulation on anesthetized rat models. We will characterize the cell-type specific neural responses to tFUS stimulation in somatosensory cortical circuits using multi-channel electrophysiological recordings in an in vivo anesthetized rat model. We will further increase the precision of our interrogations through cell-type specific optogenetic rat models. Aim 2. Investigation of intrinsic in vivo cell-type specificity of tFUS in awake head-fixed rats. Uninhibited by anesthesia effects, the awake head fixed model is ideal for the investigation of tFUS neuromodulation on the spatial and temporal activation of different cell types, as well as the propagation of brain activities across local neural networks in the awake brain. Aim 3. Frequency specific modulation of tFUS to induce plasticity in anesthetized and awake head-fixed rats. We will systematically study the long-term effects of tFUS stimulation on synaptic connectivity. We will test the hypotheses that 1) tFUS stimulation is able to encode frequency specific information inducing sustained synaptic plasticity in the hippocampus, and 2) the pattern of the tFUS stimulation parameters has a significant effect on the degree of change. The successful completion of the proposed research promises to uncover the in vivo cellular mechanism of tFUS by investigating in vivo cell- type specific responses to ultrasound stimulation at somatosensory cortex and the induction of long-term effects at both the hippocampus and somatosensory cortex. We will systematically characterize, model, validate and understand the in vivo neuronal and inter-neuronal responses to tFUS stimulation, not only to propel the translation of neuromodulation therapies to clinical utility but also further the understanding of the specific neural circuits in healthy brains.

NIH Research Projects · FY 2025 · 2021-09

Project Summary We propose to establish an integrative graduate training program in Neural Interfacing to enhance significantly the depth and breadth of interdisciplinary education and training of the next generation of scientific and technical leaders in this important, emerging field. The proposed program is motivated by the notion that future breakthroughs in neural interfacing research will be made by engineers who innovate neural interfacing technologies with a deep understanding of the fundamental issues and principles of brain science, and who have had hands-on experience in the application of neural interfacing technology in a real-world biological setting, aiming at ultimate clinical applications. The proposed program will train doctoral students with quantitative engineering backgrounds to: 1) Develop the technological and theoretical foundations to understand brain function and dysfunction through neural interfacing, i.e., through sensing, modeling, decoding, and modulating brain dynamics at circuit, network, and whole brain levels, 2) Advance our understanding of brain function in the domains of learning, motor control, perception and cognition, and determine and quantify the adaptive changes of the brain that occur in a novel environment, when interacting with a machine, or when in an altered, diseased state, and 3) Innovate brain sensing, modeling, decoding, and modulation technologies, based both on quantifiable behavioral or perceptual outcomes and on the adaptive changes occurring in the brain during a neural intervention. Our predoctoral fellows will be trained through an array of mechanisms including a newly proposed Neuroengineering Minor, a Practicum of Neural Interfacing course consisting of 8~12 mini-lab-rotations, in-depth research rotations, thesis co-advisors, weekly training activities including seminars and colloquia, and training in career skills development and responsible conduct of research. An advisory system will help guide students through the program. Our program will lower the barriers to interdisciplinary, convergent research by training students with quantitative backgrounds to possess integrative expertise in multiple fields so that, going forward, these students can serve as catalysts for the cross-fertilization of engineering with neuroscience in the next generation. Our trainees will be well prepared for academic and industrial challenges, be competitive candidates for academic, industry and government careers, and be at the forefront of translating scientific findings and engineering and computational technology development into industrial and clinical settings. It is anticipated that trainees of the proposed training program will become future leaders through neurotechnology innovations that enhance, measure and modulate, and ultimately heal our brain through neural interfaces.

NIH Research Projects · FY 2025 · 2021-09

Advanced Spectroscopic and Computational Analysis of Metal Sites in Enzymes, Biomimetic Models, and Catalytic Intermediates. Summary/Abstract Life of all organisms, including humans, depends on the activation of small stable molecules by metalloproteins to provide selective and rapid chemical transformations. The goal of the proposed research is to elucidate how specific metalloenzymes function by monitoring and analyzing the atomic level changes that occur at the metal active site during the reaction. These studies are augmented with benchmark studies of biomimetic complexes. The understanding gleaned from this work will provide a molecular basis for finding causes and remedies of diseases. The focus of the research efforts will be on how the interfacing of the active sites in metalloproteins with the protein matrix affects enzymatic function. The primary atomic coordination to the metal is of critical importance, but in many cases, weaker secondary sphere interactions, usually hydrogen bonds from nearby amino acid residues can have significant influence as well. The investigations in our lab use advanced spectroscopic and computational methods, providing detailed characterizations of the metal active sites in proteins that can be compared with an extensive database of benchmarks that we and other researchers have gathered from synthetic model complexes over the years. To understand how an enzyme works, we study key steps in their chemical mechanism by trapping and characterizing reactive intermediates and tracking their elemental kinetics on a millisecond time scale. The work is highly collaborative as we depend on the expertise of many synthetic and biochemical research groups to ensure access to biomimetic and protein complexes that can be, through joint effort, prepared cleanly with well-defined protocols.

NIH Research Projects · FY 2025 · 2021-08

7. Project Summary Multiple early feasibility trials in humans have demonstrated that implantable Brain-Computer Interfaces (BCIs) can enable people with severe paralysis to use neural signals to control remote and digital communication technologies, including messaging and email. Such studies have demonstrated clearly that BCIs have the potential to improve the quality of life of patients who have physical disability due to paralysis of speech and upper limbs. However, until these technologies are commercialized, access to BCIs will remain limited to people involved in research studies, and only for the duration of their enrolment in the study. To address this unmet need, Synchron, Inc. has developed the StentrodeTM system, a fully implantable BCI that communicates wirelessly to an external interface on a mobile computing platform. The StentrodeTM BCI is a 16-channel array of sensors integrated into a self-expanding, stent-like substrate. The StentrodeTM is delivered endovascularly via a catheter to the Superior Sagittal Sinus, where it measures volitionally-modulated neural signals from the leg area of motor cortex in both hemispheres. A fully implantable, wireless telemetry unit digitizes and transmits the neural signals from the StentrodeTM to an external mobile processor that converts the neural signals into commands for operating a computer or other assistive device, such as a speller for communication. The Synchron team has already initiated a first-in-human trial of the StentrodeTM BCI system in Melbourne, Australia, under approvals granted by the Therapeutic Goods Administration (TGA) of Australia and the IRB of the Royal Melbourne Hospital. The first human implant was performed in a person with amyotrophic lateral sclerosis (ALS) in August, 2019. The participant has experienced no adverse events and is using the system to operate a computer and type messages to friends, family, and caregivers. The objective of the proposed research is to demonstrate in an Early Feasibility Study (EFS) that the StentrodeTM BCI communication system is safe and effective in providing a quantifiable improvement in independence and quality of life in n=6 people with severe paralysis due to ALS. Two Specific Aims are proposed: 1) Preclinical assessment of the StentrodeTM for safety and functionality to complete an FDA submission, and 2) Testing of StentrodeTM’s safety and efficacy in an EFS clinical trial in two centers of excellence in the USA. Under Aim 1 (UG3 phase), preclinical safety studies and software validation in large animal studies will be completed to test robustness of StentrodeTM, compliance to safety standards for Class III electromechanical implants, safety and baseline functionality in a large animal model, efficacy of custom-built software, and a functional neuroimaging study to support presurgical planning. Under Aim 2 (UH3 phase), an EFS study will test safety of StentrodeTM placement, monitoring adverse events, target patency, and device migration. When combined with eye-tracking technology, users will be trained to perform computer-based tasks using eye-gaze to control cursor position and BCI outputs to control discrete actions, such as letter or menu-item selection and zoom. Clinical efficacy outcomes will assess the restoration of independent function by use of personal devices, including technical capability (click and typing speed and accuracy, smart home, IoT, haptic feedback), independent domestic functionality (I-ADLs) and QOL and mental wellbeing (WHOQOL, MacGill QOL, HADS).

NIH Research Projects · FY 2024 · 2021-05

PROJECT SUMMARY/ABSTRACT Mindfulness-Based Intervention (MBI) training programs have been shown to reduce stress and improve a broad range of stress-related disease outcomes in initial randomized controlled trials (RCTs). For example, there is initial evidence from small RCTs that MBIs reduce symptoms in Irritable Bowel Syndrome (IBS) patients. Yet we know little about the underlying active treatment mechanisms of mindfulness training. Guided by a theoretical and conceptual model of the active treatment elements of MBIs, called Monitor and Acceptance Theory (MAT), we recently showed in two published dismantling MBI RCTs that acceptance skills training is critical for driving stress reduction effects in healthy stressed community adults. Specifically, standard MBI programs with attention monitoring and acceptance skills training were superior in reducing stress relative to parallel MBI programs that did not include acceptance skills training (or to control groups) at post-treatment. We observed consistent stress reduction effects across intervention delivery approaches in these two dismantling trials, using either a 2-week remote smartphone MBI or with an 8-week group-based Mindfulness-Based Stress Reduction (MBSR) MBI treatment approach. Here we propose the first translational trial of this mechanistic account, examining whether acceptance skills training drives stress resilience and improved symptom outcomes in IBS patients. In the largest and most well-controlled RCT of MBI training in IBS to-date (N=325), we will evaluate whether a smartphone MBI program (with attention monitoring and acceptance skills training; Monitor+Accept, MA-MBI) reduces daily life stress and IBS symptoms at post-treatment and two-month follow-up, relative to a matched MBI program with acceptance skills training removed (training in attention monitoring skills only; Monitor Only, MO-MBI) or to an active stress management training control group (Coping Control, CC). Participants will not only provide clinician and patient assessed measures of IBS symptoms at the three time points, but they will also provide sensitive experience sampling assessments (using Ecological Momentary Assessment) of their stress and symptoms in daily life at each time point. Finally, as an exploratory aim, participants will provide stool samples at baseline and post-intervention to provide the first ever test of whether MBIs can alter the gut microbiome in IBS. Guided by a conceptual model, the proposed study will experimentally evaluate whether acceptance skills training is a key ingredient for stress reduction and health benefits in IBS patients, and will provide an important mechanistically-focused evaluation of the active treatment elements of MBIs for at risk stressed patient populations. We believe that this trial not only can help identify a new accessible and scalable evidence-based treatment for IBS patients, but the mechanistic focus will also help the field advance more effective and efficient acceptance skills-based mindfulness training approaches among at-risk stressed patient populations.

NIH Research Projects · FY 2024 · 2020-09

PROJECT SUMMARY The cell nucleus is a heterogeneous organelle that consists of nuclear bodies such as nuclear lamina, speckles, nucleoli and PML bodies. These structures continuously tether and tug chromatin at the small and large scales to synergistically orchestrate dynamic functions in distinct spatio-temporal compartments. A major obstacle to the production of navigable 4D reference maps and relating structure to function in the nucleus remains understanding how these different scales of organization influence each other. In particular, we have a poor understanding of the large-scale genome organization. Growing evidence suggests that such nuclear compartmentalization is causally connected with vital genome functions in human health and disease. However, the principles of this nuclear compartmentalization, its dynamics during changes in cell conditions, and its functional relevance are poorly understood. One lesson from Phase 1 4DN was the huge gap in throughput between imaging methods, that directly measure large-scale multi-landmark relationships, and genomic methods, that aim for whole genome high-resolution maps but are indirect measurements and provide limited information about large-scale compartments. For this 4DN UM1 Center application, we propose to meet these needs through the following Aims: (1) Generate multi-modal imaging and genomic datasets to reveal the structure, dynamics, and function of nuclear compartmentalization; (2) Develop and apply computational tools for data-driven genome structure modeling and integrative analysis of nuclear compartmentalization; (3) Develop an integrative analysis and visualization platform with navigable 4D reference maps of nuclear organization. The combined datasets and results of our proposed approaches will advance our understanding of nuclear compartmentalization, the interwoven connections among different nuclear components, and their functional significance. Our new integrative analysis tools and data-driven predictive models will produce more complete nuclear organization reference maps that integrate large-scale chromosome structure data from live and super-resolution microscopy with multi-modal genomic data including smaller scale chromatin interaction maps and predict functional relationships and dynamic responses. Our navigable reference maps will be publicly accessible through an analysis platform that provides interactive visualization of multiple data types, thus enabling investigators with diverse expertise to simultaneously explore their own data and related datasets/tools and promoting collaborations that will open new horizons into the role of the 4D nucleome in human health and disease.

NIH Research Projects · FY 2025 · 2020-09

PROJECT SUMMARY / ABSTRACT The most pressing issue with hearing aids and cochlear implants is that they function poorly in noisy environments for most users, where even mild hearing loss can make it difficult to ignore background sound. Suppression of unwanted sound is crucial for communication in social settings, such as the workplace. Inability to understand speech in these situations, called masking, can lead to social isolation and reduced employment. Two principal types of masking interfere with optimal function of hearing aids and cochlear implants. The first type, called energetic masking, is well characterized through psychophysics, physiology and modeling. The second type, called informational masking, is currently only defined as a psychological construct and much less understood. Energetic masking occurs when target speech and background sound excite the same auditory nerve fibers at the same time. Even an ideal listener would be mostly unable to recover an energetically masked target. In contrast, informational masking occurs even when target and background sound do not overlap in time and frequency, and when an ideal listener could restore the target information. Informational masking thus holds a key to improved hearing aid and cochlear implant design. Moreover, individual listeners differ dramatically in their ability to suppress informational masking. However, hearing aids and cochlear implants only intend to mitigate energetic masking, ignoring vulnerability to informational masking. Towards improved fitting of hearing aids and cochlear implants, we propose to develop an objective scale of vulnerability to informational masking based on cortical function. We propose to examine cortical mechanisms of informational masking in humans and in an animal model organism of human auditory processing, the Mongolian gerbil (Meriones unguiculatus). First, we will test normal-hearing human listeners as well as gerbils under conditions of informational masking and simultaneously record from auditory cortex. In humans, we will record the hemodynamic response of blood oxygenation, using a quick and robust assessment technique with clinical relevance. In gerbils, we will measure neuronal activity in auditory cortex from trained animals. We will use this data to develop an objective metric of an individual’s vulnerability to informational masking. Second, we will examine the neuronal mechanisms of informational masking by introducing rapid unpredictable changes in background sound and assessing if high vulnerability to informational masking is due to predominant reliance on suppressing background activity (as opposed to enhanced responses to the target) in humans and gerbils. Third, using our animal model, we will test how hearing loss affects susceptibility to informational masking. Collectively, this proposal will functionally define informational masking at both perceptual and cortical processing levels. The results are expected to significantly advance our understanding of the origins and scope of this central auditory processing deficit in common everyday situations with background sound.

NIH Research Projects · FY 2024 · 2020-09

Abstract: All cancer cells need to maintain telomere length for immortality. While most cancer cells reactivate telomerase, a reverse transcriptase, to elongates telomere from an RNA template, about 10-15% of cancer cells are telomerase-negative and adopt a homologous-recombination based alternative lengthening of telomeres (ALT) pathway. ALT cells exhibit many abnormalities in nuclear organization, including the formation of nuclear bodies called APBs for ALT telomere-associated promyelocytic leukemia nuclear bodies, clustering of telomeres within APBs, and the formation of RNA foci on telomeres with a long non-coding RNA called telomere repeat-containing RNA (TERRA). These unique features are used as biomarkers for ALT diagnosis and can be attractive therapeutic targets because of reduced side effects on healthy cells that do not share these features. However, how these features contribute to telomere maintenance and ALT cancer cell growth remain elusive, due to the lack of conceptual model as well as experimental tools to monitor and control their assembly and function in live cells. Based on our observation that APBs exhibit liquid behavior and long non- coding RNAs can phase separate with RNA-binding proteins, we propose a liquid-liquid phase separation model for the assmembly and function of these ALT specific features. We hypothesize TERRA phase separates with its interacting proteins to nucleate APB liquid droplets. The liquid nature of APBs droplets (also called condensates) would promote coalescence of APBs to drive telomere clustering. Meanwhile, condensation of APB droplets can concentrate DNA repair factors, providing opportunities for telomeres to use one another as repair templates to elongate within APBs. To test our hypothesis, we developed a state-of-the- art optogenetic approach to control APB assembly. We demonstrate that liquid phase separation underlies APB assembly and coalescence of APB droplets indeed drives telomere clustering. Building on our ability to control telomere clustering and APB assembly and by collaborating with experts in super resolution microscopy, nuclear mechanics, chromosome organization and ALT cancer, we will investigate how DNA repair factors are recruited to and organized in APB condensates for ALT telomere DNA synthesis (Aim 1) and how telomere clustering leads to unique genome organization and gene expression in ALT cells (Aim 2). We will then extend our optogenetic tools to control RNA and dissect TERRA contributions in ALT (Aim 3). Results obtained by manipulating cultured ALT cells will be confirmed by characterizing ALT tissue or creating de novo ALT phonotypes in primary human cells. Our results will provide mechanistic understanding on how protein and/or RNA phase separation contributes to ALT cancer, which will offer the potential to develop strategies specifically targeting these unique phase separation processes, rather than the existing molecules that shared by heathy cells, for ALT cancer treatment. Such therapies are also beneficial for treating telomerase-positive cancer as these cancer cells can escape telomerase inhibition and adopt ALT for telomere maintenance.

NIH Research Projects · FY 2024 · 2020-09

Project Summary: In this research project, we develop synthetic inorganic copper complexes to understand the fundamental aspects of structure and function in Cu-dependent monooxygenase enzymes. These metalloenzymes contain 1 or 2 Cu ions in their active center and they couple the reduction of O2 with the oxidation of substrates via formation of transient Cun/O2 species. We are particularly interested in studying the reactivity of mononuclear Cu/O2 intermediates since they have been proposed as active oxidants in the hydroxylation of strong C-H bonds in enzymes such as particulate methane monooxygenases (pMMOs) and lytic polysaccharide monooxygenases (LPMOs). Many questions concerning the identity of the active Cu/O2 species remain unanswered, including: i) oxidation state of Cu (CuI vs. CuII vs. CuIII); ii) reduction/protonation state of O2 (O2Ÿ−,(H)O22−, (H)O2−) and the pKa and redox potentials associated with these Cu/O2 species; iii) mechanism by which the Cu/O2 intermediates carry out C-H hydroxylations (e.g. O-O cleavage mechanism before or after C-H oxidation?; generation of high-valent Cu-oxyl species before substrate hydroxylation?). In this research proposal, we tackle this problem using two different approaches: 1) We utilize ligand scaffolds (L) that contain C-H substrates covalently attached to their structure (substrate- ligands) that permit us to generate and characterize LCu/O2 species and evaluate their reactivity towards intramolecular C-H hydroxylation. Substrate-ligand modifications will permit us to: i) evaluate the ability of the Cu/O2 species to oxidize sp3 C-H bonds and sp2 C-H bonds; ii) control the stereo-electronic properties of the Cu complexes by the use of different ligand donors (i.e. N2, N3, N4) that will lead to the generation of mononuclear and dinuclear LCu/O2 species, and analyze their reactivity towards intramolecular C-H hydroxylation including characterization of reaction intermediates, kinetics and computations; iii) utilize this approach (Cu-directed hydroxylations) to develop synthetic protocols to promote challenging organic transformations such as enantioselective C-H hydroxylations and one-pot synthesis of 1,3-oxazines. 2) We synthesize mononuclear Cu complexes bearing redox-active ligands with tunable H-bonds that stabilize Cu-hydroxo and Cu-oxyl cores. These unusual Cu complexes are able to reach multiple oxidation states via oxidation of the metal and/or ligand scaffold. These high-valent CuO(H) cores will be characterized by various spectroscopic methods and their ability to perform biorelevant intermolecular 2e− C-H hydroxylations will be examined systematically using the Bordwell equation (i.e. species with higher redox potential and higher pKa should be capable of oxidizing stronger C-H bonds), kinetic experiments and analysis of the reactions products derived from hydroxylation (e.g. organic product(s) and oxidation/protonation state of the final Cu complexes). Overall, these studies will contribute to a broader understanding of the biochemical role of Cu ions involved in O2 reduction and biologically relevant oxidations.

NIH Research Projects · FY 2024 · 2020-08

Project Summary Optogenetics is a powerful tool for relating brain function to behavior because it enables cell- type specific manipulation of neurons with millisecond temporal precision and artifact-free neural recordings. Such capabilities are particularly needed in studies using non-human primates (NHPs), where sophisticated behavioral techniques are commonly employed but neurophysiological tools have lagged those used in other model species. While the use of optogenetics in NHPs has grown rapidly in recent years, the full power of the technique requires the ability to perform large-scale, bi-directional study of neural circuits. Systems to achieve this have become widely used in other animal models, particularly mice, while there have been limited systems implemented in NHPs. In this proposal, a large-scale, high-density, and stable optoelectric neural interface (smart dura) for large brains will be developed and validated in macaques, for the first time. This novel interface enables simultaneous electrical recording from 4096 electrodes and optical stimulation in 4096 sites over about 5 cm2 of cortex, which is more than two orders of magnitude higher than the state-of-the-art technology. As opposed to existing surface electrocorticography (ECoG) electrode arrays, the proposed neural interface is in the form of an artificial dura that monolithically embeds electrical recording and optical stimulation functionalities such that it can permanently replace the native dura as a chronic, seamless neural interface, while maintaining the natural cranial pressure. Therefore, this novel design combines the best of passive/static artificial dura windows and functional surface electrode arrays in one unified platform. The proposed smart dura enables long-term recording, provides new opportunities for creating sophisticated closed-loop stimulation and recording paradigms, and advances the development of new stimulation-based therapies. The smart dura can be implanted as a stable port into large brains and consists of high-density recording electrodes as well as optical micro light sources all embedded in a hybrid biocompatible polymer platform. In this project, a novel fabrication process will be designed to implement the proposed large-scale (5 cm2) smart dura in two stages of: i) Fabricating high-density transparent electrical smart dura for electrophysiology recording and external optical access (transparent electric dura: transparent e- dura), enabled by high resolution interconnects (300 nm features). ii) The optoelectric dura (oe- dura) consisting of high-density recording electrodes and embedded micro light emitting diodes (µLEDs). In each stage of the device development, the neural interface will be tested in two hemispheres of two monkeys, with large optogenetic expression of activating opsin (ChR2) in sensorimotor cortex via electrophysiology recording, behavior, and imaging. The proposed smart dura will greatly enhance the opportunities for closed-loop optogenetic experiments in macaques, which can serve as a powerful tool for understanding brain function and for developing novel therapeutic interventions that can be translated to humans. After successful demonstration of the smart dura in this proposal, the results can be extended in future to i) develop even larger interfaces that cover the whole brain for translational use ii) integrate recording, stimulation, processing, communication and power-transfer electronics into the smart dura to enable tetherless chronic neural interfacing with freely-moving subjects.

NIH Research Projects · FY 2026 · 2020-05

Project summary/abstract In this application, we request continuation of MH123184, which aimed to understand how genetic variation alters transcription in specific cells and thereby produces psychopathology. Our research developed statistical methods to integrate single cell and tissue-level transcriptomic data. We targeted methods to identify gene communities, defined in terms of cell type and spatiotemporal window, to understand how genes act in concert to confer risk for psychopathology. We also took advantage of an exciting new avenue of research to approach these challenges, namely CRISPR screening. This innovation has emerged as a powerful tool to characterize the effects of genetic perturbations on the entire transcriptome at a single-cell level. Here we propose research covering three related themes, all of which capitalize on CRISPR advancements: (1) develop powerful and well-calibrated tests for the effect of CRISPR perturbations on gene expression by inferring latent factors; (2) develop methods for removing the effect of unmeasured confounders in high throughput screens; and (3) develop methods for imputation and denoising for multiomic data that facilitate downstream testing of omic readouts. Each of these aims is motivated by pressing needs in the field. First, due to small samples and the sparsity of the response variable, it is essential that we enhance the power and interpretability of CRISPR tests by accounting for co- regulation and convergent function of genes. Aim 1 achieves this purpose by estimating latent factors that represent co-regulated genes and by inferring a similarity matrix among gene perturbations. As CRISPR screens advance to more biologically complex settings, such as model organisms, unmeasured confounders will play a more important role, and new methods are needed to control for these effects. Aim 2 develops two approaches to this challenge: an innovative use of negative control variables, as motivated by the causal literature, and key advances to the classic surrogate variable analysis method. For the field to move toward efficient use of multiomic data, data derived from multiple sources will be required. These resources will invariably have missing data. Methods to account for imputation of missing data are needed. Tools developed for variational autoencoders show great promise; however, as described in Aim 3, they need to be paired with semiparametric inference tools to ensure robust and well calibrated downstream analysis. By applying what we learn from these three aims to available resources, most from distributed resources and some from our collaborations, we expect to shed more light on the neurobiological mechanisms of mental illness. We are well positioned to move between theory and data because we have a diverse team of investigators lead by the PI (Roeder), who has decades of experience in statistical genomic field and co-investigators Wasserman and Lei, who are experts in theory and methods for high dimensional and causal inference. The proposed research fits Goal 1 of NIMH’s Strategic Plan, advancing basic science of brain, genomics, and behavior to understand mental illnesses.

NIH Research Projects · FY 2025 · 2020-03

Pain is among the most pervasive and universal forms of human distress. Pain typically is measured by patient report or clinician impressions, either through clinical interview or the visual analog scale. However, patient reported pain is difficult to interpret and in some circumstances not possible to obtain. To advance pain assessment, monitoring, and intervention, we propose (1) a savvy technology based on automatic facial, head, and body movement analysis for a reliable and valid assessment of the occurrence and intensity associated with five causes of acute and chronic low back pain (LBP); (2) inform our understanding of indicators of chronic LBP. Participants' face, head, and body movement will be recorded during clinical assessment using high-definition digital video cameras during extension, flexion, and rotation movements. The obtained video-recordings, taken during a first visit to the clinic and 3 follow-up visits, will be used for the development of automatic measures of the occurrence and intensity of pain. To investigate the generalizability of the proposed automatic measures, we will explicitly train and test the proposed classifiers on five different types of acute and chronic LBP. To do so, face, head, and body movement will be automatically tracked using our fully- automatic methods. The tracking results will be used to train end-to-end deep-leaning based classifiers to automatically measure the occurrence and intensity of LBP. To investigate the validity of the proposed classifiers, we will compare automated measurement to the patient- and clinician- rated visual analog scale, brief pain inventory, and continuous observer ratings of pain intensity from the video recordings. MANOVA will be used to quantify the relationship between the individual modalities and their combination for the measurement of the occurrence and intensity of the five LBP conditions and for chronic and acute conditions. To inform our understanding of how LBP evolves into a chronic form, we will use Ecological Momentary Assessment (EMA) to collect behavioral and contextual information beyond the video-recordings and pain scores' assessments. Participants will be monitored for 6 months, at a frequency of 7 consecutive days per month (1 week per month), and 4 prompts per day, to identify those who evolved to chronic LBP. EMA measures will be used to investigate whether pain intensity differs both between and within LBP groups and investigate factors that are associated with the development of chronic LBP.

NIH Research Projects · FY 2025 · 2019-12

PROJECT SUMMARY/ABSTRACT The broad aim of this research is to evaluate whether a brief and cost-effective value affirmation writing intervention reduces physical symptoms, stress and medication non-adherence among breast cancer patients taking aromatase inhibitors (AIs). The majority of postmenopausal women diagnosed with breast cancer will be prescribed an AI; and while effective, these medications lead to toxic physical symptom side effect profiles and poor adherence. Interventions that target these side effects and the stress they cause may lead to improved medication adherence and therefore improved survival rates for breast cancer patients taking AIs. In pilot work from our group and others, thinking or writing about important personal values in value affirmation interventions have been linked to fewer physical symptoms, decreased stress, and improved medication adherence. While promising, these initial studies had small samples or were cross sectional in nature, and did not evaluate potential neurobiological mechanisms for intervention effects. The proposed study will be a randomized controlled trial (RCT) comparing a 6-month, portable value affirmation writing intervention (N=125) to a control writing intervention (N=125) in breast cancer patients taking aromatase inhibitors. Participants will complete self-reported measures of physical symptoms, adherence and stress at baseline, post-intervention, and 6-month follow-up. Medication adherence will also be assessed via electronic pill bottle reporting and refill latency during the study period. As an exploratory mechanistic aim, a subsample of AI patients from the parent RCT (N=120, 60 per intervention arm) will complete their mid-intervention value affirmation or control writing task while undergoing functional MRI. Thus, the proposed study will test two specific aims and an exploratory aim to determine: if value affirmation writing reduces physical symptoms and stress (Aim 1), improves medication adherence (Aim 2), and if neural reward activation during value affirmation is a neurobiological mechanism for these benefits (Exploratory Aim 1). The proposed RCT has the potential to identify a powerful, cost-effective, and easy-to-implement value affirmation intervention for reducing symptoms, stress, and medication non-adherence in breast cancer patients taking AIs.

NIH Research Projects · FY 2023 · 2019-09

ABSTRACT: There is a critical need for Influenza forecasting among public health decision makers, large organizations, healthcare participants, and the general public. We focus in this proposal on the first and last categories. For federal and state officials, influenza forecasting can help inform the timing of critical communications, vaccination campaigns, messaging to state and local public health agencies, hospitals, healthcare professionals, and the public. During flu pandemics, forecasting can inform the formulation and execution of strategies for vaccine development, vaccine distribution and (application), dissemination of antivirals, and recommendations for non-pharmaceutical interventions. Further, public health communications to healthcare providers can result in more informed doctors’ decisions regarding use of antibiotics and hospitalization. For the general public, reliable short term forecasts can increase the credibility of and trust in pubic health authorities, resulting in greater adherence to recommendations. Short term forecasts can also inform individuals’ decision making, especially with regard to behavior and exposure of flu-vulnerable populations: the elderly, the very young, the pregnant and the immunocompromised. Members of these groups may reconsider their travel plans and other public exposure if warned in advance of an impending epidemic wave at a specific location. While such decisions might also be informed by existing flu surveillance, inherent latencies in traditional surveillance means that nowcasting and short-term forecasting provide a few weeks advance notice -- enough to save many lives. To advance the state of the art in influenza forecasting, we propose an Influenza Forecasting Center of Excellent that will enable and improve the usefulness of forecasts of both seasonal influenza and pandemic influenza, to inform public health responses and policy development at the national, regional, and state level. The center will (1) review and, if needed, revise existing forecasting guidance, targets, and accuracy evaluation, at the national, regional, and state levels; (2) refine methods to create forecast ensembles; (3) identify methodologies and data sources that increase forecast accuracy for start and peak week forecasts, peak intensity, and short-term forecasts at the national, regional, and state level; (4) develop communication products and data visualization methods to describe forecast results and uncertainty for federal and state public health officials and the public; and (5) develop and adapt successful seasonal methodologies, data sources, and communication approaches for forecasting the timing, intensity, and short-term trajectory of an emerging influenza pandemic.

NIH Research Projects · FY 2025 · 2018-09

Project Summary/Abstract Iron and 2-oxoglutarate-dependent (Fe/2OG) enzymes, representing a superfamily of non-heme mononuclear iron-containing (NHM-Fe) enzymes, have garnered strong research interests from fundamental enzyme mechanism studies to bioengineering/biocatalysis explorations in recent years due to their exceedingly diverse catalytic reactivities and simple enzyme architectures. Radical halogenation reactions via C-H bond activation catalyzed by Fe/2OG halogenases are particularly attractive for chemical synthesis and biocatalysis applications, since these enzymes can install carbon-halide bonds in a regio- and stereo-specific manner, a feat that has yet to be achieved by organic synthetic methodology. As revealed by the mechanistic studies of carrier protein- dependent Fe/2OG halogenases, the key step in the radical halogenation mechanism is the selective halide radical transfer from the hydroxo-Fe(III)-halide intermediate to the substrate radical generated by the key reactive species, the ferryl (Fe(IV)=O) intermediate. However, a consensus mechanism to explain the selective halide transfer in Fe/2OG halogenases has not been reached, particularly the controlling factors to avoid hydroxyl radical transfer to lead to hydroxylation reaction are not fully revealed. Additionally, the reasons why Fe/2OG enzymes cannot perform fluorination reaction are completely unknown. In this project, we will bridge these knowledge gaps by studying two newly discovered carrier protein-independent Fe/2OG halogenases that catalyze chlorination reactions to generate halogenated nucleotide natural products and halogenated free- standing amino acids. By using an integrative approach consisting of mechanistic probe design and synthesis, enzyme product structural determination via LC-MS and NMR analysis, transient enzyme kinetics, advanced spectroscopic characterization and molecular dynamics simulations, we will elucidate the influence of protein- substrate interactions and dynamics in controlling efficient halogenation, explore the effect of different iron-bound anions (e.g. Cl- vs. F-) to the electronic structure and the reactivity of the ferryl intermediate, test new chemical strategies to enable fluorination in Fe/2OG enzymes, and expand the substrate scope of these enzymes for potential synthetic applications. Given the importance of halogen-containing organic molecules in the modern pharmaceutical and agrochemical applications, mechanistic elucidation of these newly discovered halogenases will lay scientific foundation for future biocatalytic applications of these unique enzymes.

NIH Research Projects · FY 2026 · 2017-09

PROJECT SUMMARY Brain image data collected by the neuroscience community will continue to grow dramatically in the next 5-10 years as the push to image physically larger volumes, such as whole brains from humans and other primates, becomes a focus. The scientific community, with the support of NIH data sharing policies, expect this valuable data to be shared and made available to the community as rapidly as possible. The Brain Image Library (BIL) is a public resource serving the neuroscience community by providing a persistent scalable repository for sharing microscopy data generated by brain researchers. Importantly, raw data can be archived with BIL, and all post-processing and analysis can be completed at BIL and rapidly made publicly available. BIL maximizes data reusability by providing rapid access to a range of community-specific software suites and user- specific software which can quickly access a variety of high-performance computational resources including GPU, CPU, high-RAM and machine learning nodes. This one-stop access encourages the reuse of data by making it easily accessible to a variety of scientists and promotes transparency and scientific discovery as the outcomes of any meta-analysis can be made public and tied directly to the original data content.

NIH Research Projects · FY 2025 · 2016-06

Project Summary The long-term goal of the research program is to develop and establish a novel electrophysiological source imaging technology to localize and image epileptogenic brain tissues aiding pre-surgical planning in focal (partial) epilepsy. Epilepsy is a common neurological disease impacting more than 3.4 million patients in the US and 70 million globally. The standard clinical routine heavily relies on using intracranial EEG (iEEG) implanted into the brain to determine seizure onset zone, to aid in the localization of epileptogenic zone (EZ), despite the limited coverage of iEEG electrodes and invasive nature of the multiple-day procedure. There is a clinical need to develop a noninvasive neuroimaging approach based on electrophysiological recordings that can reliably image and delineate the EZ from relevant epilepsy biomarkers. In this research project, we propose to establish a novel unsupervised machine learning framework that will be able to process unmarked, continuous, and long-term EEG recordings of focal epilepsy patients to detect high frequency oscillations (HFOs), automatically identify the pathological HFOs (pHFOs, i.e., HFOs riding on spikes), and localize and image epileptogenic brain activity from the identified pHFOs. The proposed techniques will be rigorously validated in over 120 focal epilepsy patients against clinical findings from iEEG recordings and surgical resection outcomes. Our specific aims are: Aim 1. Development and evaluation of a novel unsupervised machine learning technique to identify pathological HFOs from continuous EEG recordings. This aim will establish a novel unsupervised machine learning approach for automatically identifying pHFOs originated from epileptogenic activity, by incorporating data-driven feature learning methods. Aim 2. Development and validation of a novel brain tensor decomposition imaging approach for HFO source imaging. We will develop a novel source imaging framework imaging epileptic sources in temporal, spectral, and spatial domains. This aim will establish a novel data-driven source imaging approach for accurate mapping and localization of EZ from scalp recorded pHFOs. Aim 3: Validation of scalp-detected pHFOs by iEEG-identified pHFOs, with clinical findings of EZ in focal epilepsy patients. We will test the hypothesis that the scalp-EEG identified pHFO events reflect the essential features of iEEG-identified pHFOs, and that both are indicative of the EZ determined from clinical iEEG and confirmed by surgical resection outcome. This aim will establish the relationship between scalp-identified pHFOs with iEEG-identified pHFOs and the underlying epileptogenic networks. The successful completion of the proposed research will establish a novel unsupervised machine learning technology to detect and identify pathological HFOs from continuous and long-term scalp EEG recordings, and localize and image the underlying epileptogenic zone noninvasively and accurately. The establishment of such novel technology promises to significantly improve the clinical management of focal drug- resistant epilepsy, which is currently limited, benefiting numerous patients and the healthcare system.