Clemson University

NSF Awards · FY 2025 · 2025-10

Whether entering a password, using FaceID, or confirming a push notification, people authenticate (prove their identity) to access accounts and information. While authentication tools like passwords, multi-factor authentication, and biometrics help keep data secure, their design and frequency can interrupt important tasks and cause people time, effort, and frustration. These burdens can lead users to take shortcuts that make authenticating easier but threaten security, such as reusing passwords across accounts. This project addresses the human side of authentication by examining how often, when, why, and how people authenticate and the challenges they face while doing so. By identifying common patterns and barriers across contexts, including contexts that are key to the future success of AI, this research informs the design of authentication systems that are more usable, efficient, and aligned with users' daily lives. The research insights will ultimately support US national interests by strengthening cybersecurity and enhancing access to information and services by making systems more secure and easier to use. This project takes an interdisciplinary approach to improve authentication by focusing on human factors. Little systematic research has been carried to understand how often, when, and how people authenticate, and the associated personal and professional costs of authentication. By evaluating authentication across domains and contexts, this project produces novel, authentication-tracking tools to obtain detailed metrics about users’ daily authentication patterns and assesses cross-cutting themes from two domains: traditional large organizations and microtask crowdsourcing platforms (which underpin progress in AI). An annotated database of observed authentication events, combined with surveys, time diaries, and interviews, helps identify usability challenges, interruptions, and associated risks. Findings are applied to propose future directions for forward-thinking authentication systems that minimize costs, enhance usable security, and consider how to preemptively address threats posed by advancements in AI. These insights promote a more trustworthy cyberspace and support national cybersecurity. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-09

NSF Awards · FY 2025 · 2025-09

An award is made to Clemson University to enable a five to ten-fold increase in the number of different things that can be measured in a mixture simultaneously (multiplexing), using widely available, affordable, and reliable methods. This project will have broader impacts related to enhanced scientific access to highly-multiplexed experimental capabilities. It will provide training for the next generation of scientists by engaging undergraduates and high school students from a variety of disciplinary backgrounds and by mentoring doctoral students to carry out the work. Findings will be integrated with courses at Clemson University and with continued implementation of new student learning assessment methods. They will be available by open-source access of results, protocols, raw data, and analysis tools, and by public outreach through the Clemson University student chapters of scientific societies. Fluorescent reagents are a workhorse across the biological sciences. Conventional fluorescence multiplexing is limited to about four colors simultaneously, with recent advances reporting from 20 to 40 colors. This project will enable about 200 colors, which is a five-ten-fold increase. The approach is compatible with other current methods, which could expand multiplexing another five-ten-fold, to between 1,000 and 2,000 measurements. Increased measurement multiplexing usually enables breakthrough advances in biological understanding. The recent availability of compatible and affordable hardware will drive applications and user communities. Developmental and tissue biology questions could be probed with unprecedented depth by high-dimensional profiling. Overlaying high-dimensional data from this research in a spatial tissue context will synergize with ongoing efforts to build comprehensive maps of every single cell in the body. Microbial communities from skin, gut, soil, or ocean could be stratified, isolated, studied and combined with unprecedented precision. The generality of fluorescence combined with assay accessibility will enable wide adoption by user communities in multiple disciplines to answer novel biological and other scientific questions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Collaborative Research: Research Initiation: Mixed Methods Study of Rural Engineering Students' Sense of Belonging at a Midwestern Research-Intensive University This project aims to serve national priorities by initiating research on the professional formation of engineers and supporting pathways into and through engineering for students with varied backgrounds, interests, and experiences. Meeting evolving workforce needs requires not only more engineers but professionals whose perspectives reflect a spectrum of experiences, including those shaped by rural communities. Research universities play a vital role in cultivating the next generation of engineering talent. However, fostering environments where students can thrive requires more than simply providing access—it also requires a sense of belonging. For many, feeling connected is essential to staying engaged and persisting in their studies. When that connection is lacking, the risk of leaving the field increases, limiting opportunity and reducing the breadth of ideas that drive innovation. Despite this, little research has examined how rural engineering students experience belonging at large universities—a gap this project will address. This National Science Foundation Research Initiation in Engineering Formation (RIEF) award to the University of Wisconsin–Madison and Clemson University will explore rural engineering students’ sense of belonging and identify opportunities to strengthen connectedness. Using surveys followed by in-depth interviews, the research will gather rich, first-hand accounts to contribute new knowledge about varied experiences in engineering education. These insights will inform efforts to improve retention and help ensure that talented students from all communities can advance into engineering careers. This study aligns with the goals of the NSF RIEF program by advancing innovative research on engineering formation while strengthening the capacity of early-career faculty to lead impactful educational studies. Ultimately, the findings will support a more resilient engineering workforce enriched by a broad range of perspectives. The goals of this research are to explore rural engineering students’ sense of belonging, identify opportunities to improve connectedness, and develop monitoring programs and targeted interventions for students at risk of leaving the engineering pipeline. The study will address three research questions: (1) What factors influence a student's self-identification as rural and how strongly do they identify with their rural background or experiences? (2) In what ways does this identification shape their college experience, environment, and sense of belonging? (3) How can institutions better support students in strengthening their sense of belonging earlier in their college journey? A mixed-methods approach will begin with a quantitative survey available to all students. All students will be included and participants will indicate whether they consider themselves to have rural backgrounds or experiences, and those who self-identify as rural will be included in the analysis. Allowing students to define rurality in their own terms will yield nuanced insights into how rural identity relates to their experiences. Survey findings will guide the selection of participants for follow-up semi-structured interviews to gather in-depth qualitative data. Analysis will draw on social identity theory, ecological systems theory, and belongingness theory to build a multidimensional understanding of these factors. Expected outcomes include evidence-based strategies and tools to help institutions create supportive environments, identify students at risk of disengagement, and implement targeted interventions. More broadly, this work will inform and support efforts to expand opportunities for aspiring engineers by deepening understanding of how identity influences pathways into and through engineering education. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Reinforcement Learning (RL) is a machine learning paradigm that strives to make optimal decision-making based on experience acting in an environment. In many cases, the "environment" refers to a simulator in the training stage and refers to the real world in the deployment stage. Training in the simulator brings a lot of advantages: lower cost, more safety, and more flexibility. However, it is almost impossible to design a perfect simulator that is identical to the real world. Thus, a decision-maker trained in the simulator may not function well in the real world. The discrepancy between the simulator and the real world is called the simulation-to-reality (sim-to-real) gap. This project will build new technologies to close the sim-to-real gap in both the training and the deployment stages. The research outcomes will benefit the development of next-generation RL techniques, which can improve the availability, applicability, and generalization of RL, and minimize the gap of RL between common practices and real-world practices. This project proposes to close the sim-to-real gap in reinforcement learning by three mechanisms: randomization, alignment, and derivation. Specifically, 1) the randomization mechanism generates a set of homogeneous simulators by original simulator parameter randomization. The simulator set will cover a wider range of state-action regions than the original simulator, have a larger overlap with the real-world environment, and thereafter result in a smaller sim-to-real gap. This mechanism is especially useful when the sim-to-real gap is large and the simulator is only accessible for training the simulator-optimal policy, but not accessible during the sim-to-real transfer process. 2) The alignment mechanism makes the simulator more like the real world during the transfer process. The alignment mechanism not only closes the sim-to-real gap but also is low-cost and high-efficiency, thus, accelerating the transfer process. This mechanism is especially useful when the sim-to-real gap is relatively small and the simulator is accessible in both simulator-optimal policy training and sim-to-real transfer. 3) The derivation mechanism directly derives an optimal policy from real-world offline data without any simulator. It first estimates state-action values from offline data and then derives the policy by function approximation. This mechanism is especially useful when offline data has been collected, but the real-world dynamics are unknown so it is unlikely to build a faithful simulator. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Electrostatic free energy (EFE) calculations are indispensable for the quantitative analysis of biological processes, as they characterize the polar interactions between charged biomolecules such as proteins, DNA and RNA, and their surrounding ionic solvent environments. As one of the most widely used implicit solvent models, the Poisson-Boltzmann (PB) model computes EFE as the difference between PB energies of biomolecules in two reference states, typically vacuum and solvent. The success of classical PB theory relies on two restrictive assumptions: (i) biomolecules remain rigid and do not change shape when moving between states, and (ii) identical computational procedures are used for both states. However, under physiological conditions, proteins are inherently flexible and undergo conformational changes during solvation and binding. This project aims to overcome the rigidity limitation by introducing a generalized PB theory that accommodates non-rigid biomolecular structures. This enables PB models to handle shape changes in key biological processes such as solvation and binding. The proposed algorithms will be implemented in DelPhi, an open-source PB package, and they will be applied to other popular PB solvers in the form of post-processing patches. The new computational tools will be distributed free of charge to academic users, making them accessible to the broader biological research community. In addition, this project will provide interdisciplinary research and training opportunities for undergraduate and graduate students in biophysical modeling, computation and mathematical analysis. Outreach and dissemination activities will be developed to engage broader audiences and foster public understanding of how computational science contributes to human health and biomedical innovation. The limitations of the classic PB theory essentially stem from the fact that EFE computed by the PB model involves self-energy terms, that is, the singular charge at an atom center will interact with the potential induced by itself, which yields infinite energy values at each atom center. The rigidity assumption and identical numerical discretizations in two reference states enables the cancellation of these infinitely large self-energies terms between states. This project introduces a novel partition of the PB energy functional to separate the singular self-energies from the regular parts, so that the self-energy difference due to conformational changes can be analytically formulated and is free of singularities in subsequent numerical computation. This approach applies to both sharp-interface and diffuse-interface PB models, and it employs distinct strategies for regularization and non-regularization approaches in the numerical treatment of the PB equation. The proposed PB theory represents the first computational method capable of accurate EFE prediction for non-rigid biomolecules. This innovation provides a more precise physical modeling of solvation and binding processes and yields more accurate polar solvation and binding energy predictions. The proposed research will have a broader impact to the field of molecular biosciences, by providing improved binding energy estimations to several PB applications in drug design and mutation predictions. This project is jointly funded by the Mathematical Biology Program in the Division of Mathematical Sciences and the Chemical Theory, Models and Computational Methods Program in the Division of Chemistry. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This BRIDGE-QC training project addresses the nationwide shortage of professionals in Quantum Information Science (QIS) and the limited regional presence in this rapidly emerging field. QIS is advancing quickly, with transformative applications in sensing, computing, and networking. However, the United States faces a critical workforce gap in QIS and has called for intensified efforts to attract, educate, and develop a robust quantum workforce. Despite these national efforts, the educational infrastructure and research ecosystem needed to train quantum scientists remain nascent. BRIDGE-QC aims to meet the national demand for a quantum-ready workforce while also expanding opportunities in the field. BRIDGE-QC has two primary goals: (1) to develop a highly skilled quantum computing workforce by training students as quantum users, researchers, and contributors; and (2) to build and sustain a talent pipeline. The project takes a hands-on, interdisciplinary, and collaborative approach, creating a dynamic learning environment that combines formal coursework with informal, student-driven experiences. Core activities include new course development: Introduction to Quantum Computing and Quantum Algorithms and Applications, and the establishment of a quantum computing minor open to students across disciplines. To foster early engagement, BRIDGE-QC recruits high-achieving sophomores and juniors through Clemson's Creative Inquiry program and the Undergraduate Research Experience, and supports student-led initiatives such as the Quantum Club and Quantathons to promote community and partnerships with industry and research institutions. The project will deliver a comprehensive set of educational materials, including lecture notes, labs, onboarding modules, tutorials, source code, and scholarly publications. These resources will be publicly accessible and disseminated to support broad adoption at the state, regional, and national levels. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

In plants, like corn, leaf aging (also known as senescence) is accompanied by regular cellular and physiological changes. Some of these changes are outwardly visible--like drying, loss of green color due to reduced photosynthesis, and browning under drought. Other changes are not obvious to the naked eye but are very important for the plant. One such change involves recycling of nutrients from the aging leaf to developing parts of the plant, such as kernels on the ear. The timing of senescence is crucial—too early results in low grain yield. By contrast, plants that delay senescence and stay green longer into the growing season tend to yield more grain. Exactly what controls the timing and the many steps in senescence is not clearly known. This project will develop new ways to measure corn traits during aging, discover causative genes, and apply computational methods like artificial intelligence to understand how nutrient recycling in plants can improve agricultural outcomes. The project will also provide interdisciplinary training for undergraduate and graduate students, postdoctoral researchers, and high school teachers to help inspire and prepare the next generation of scientists, ultimately strengthening capacity in plant biology, data science, and agricultural biotechnology. This project aims to uncover the genetic, metabolic, and regulatory mechanisms that govern leaf senescence and nutrient remobilization in maize—processes that directly impact grain quality, nitrogen use efficiency, and overall crop productivity. Despite their agronomic importance, the molecular drivers of senescence remain poorly understood in cereals. To address this gap, high-resolution, time-series datasets will be generated across a genetically diverse maize panel, capturing physiological traits, metabolite profiles, transcriptomes, and chromatin accessibility. Single-cell assays will add spatial resolution to senescence-related transcriptional and chromatin changes. These datasets will be integrated using artificial intelligence tools, such as large language models, and machine learning to identify causal genes, regulatory elements, and metabolite-phenotype associations. Top candidate genes will be functionally validated using gene-editing technologies to confirm their roles in regulating senescence and nutrient allocation. The resulting insights will inform breeding and biotechnological strategies to enhance nitrogen use efficiency, reduce fertilizer inputs, and improve resilience to abiotic stress. This project is co-funded by the Division of Integrative Organismal Systems Plant Genome Research Program and by the Division of Emerging Frontiers, both in the Directorate for Biological Sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

As automation and artificial intelligence reshape our world, autonomous systems—like self-driving cars, delivery drones, and robotic assistants—must make decisions safely and efficiently in unpredictable environments. These systems operate under various sources of uncertainty, including sensor errors and unmodeled disturbances. A central technical challenge is “optimal control”: how to determine the best actions to meet performance and safety goals. However, solving optimal control problems for complex, high-dimensional systems often become computationally intractable—a challenge known as the curse of dimensionality. This research project looks to develop a novel approach to overcome this barrier by using modal decomposition techniques from linear operator theory that transform difficult nonlinear problems into simpler linear ones. This makes it easier to compute control strategies for complex tasks and high dimensional systems. The project seeks to impact technologies such as autonomous robotics and smart energy systems. It will also provide research opportunities for students from a wide range of backgrounds, institutions, and career stages through Clemson University’s programs. K–12 outreach events will promote early interest in STEM. This work aligns with national priorities in autonomy, workforce development, and innovation in control technologies. This research seeks to develop a scalable framework for optimal control of nonlinear systems under uncertainty, using the spectral theory of linear operators, especially the Koopman operator. The core insight is a formal link between Koopman eigenfunctions and the Hamilton-Jacobi (HJ) equation, a fundamental equation in control theory. This research looks to enable recasting nonlinear optimal control problems as linear ones in a transformed Koopman eigenfunction coordinate space, mitigating the curse of dimensionality. The Koopman eigenfunctions are used in the decomposition of the HJ equation into integrable and nonintegrable parts, where the integrable part is solved exactly and the nonintegrable part approximately resulting in the approximation of the HJ solution. The project intends to make novel contributions towards the computation of Koopman eigenfunctions for stochastic system based on the Feynman-Kac path integral formula. This formulation allows for scalable, data-assisted computation of optimal control policies, even under stochastic or adversarial uncertainty. The framework looks to support input constraints and extends Koopman theory into practical feedback control design. Applications will focus on controlling robotic systems, such as legged robots on uneven terrain, where conventional methods are infeasible. The project integrates algorithm development, theory, and demonstration, while also contributing to STEM education and broader access to control research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project investigates a variety of different robot locomotion that can be achieved using only fast unbalanced rotors for actuation. Preliminary experiments have demonstrated that such systems can be used to roll, jump, crawl, climb, swim, and fly. Initial analyses have connected different discrete modes of motion with special relationships between the amplitude of the rotor imbalance and the speed of the rotor spin. The project will extend understanding of these relationships, allowing the robot’s mechanical structure to be designed together with its control system, thus providing the ability to select between different mobility modes. The approach will be applied to robots made of multiple rigid components as well as robots made of soft materials. The spin dynamics of internal rotors can be easily and accurately controlled, suggesting that this approach may have practical advantages for ease of operation. The robots considered in this project have potential applications in defense, monitoring, exploration, and medicine. Inexpensive and mechanically simple robots designed in this project will be used as the basis for STEM outreach activities to high school students in upstate South Carolina. The project will realize a framework for novel multidomain mobility through a deep understanding of novel mechanics. The approach will build upon the analysis of net inertial displacements corresponding to high frequency oscillations of shape variables on a configuration manifold in the context of parametric resonance and vibrational stabilization as exhibited, for example, by the nonlinear Mathieu equation. Extensions of this framework to discontinuous systems will capture movements such as jumping. Experiments to explore potential new capabilities will include rolling-jumping robots capable of traversing unstructured and rough terrain, fast-crawling robots capable of climbing pipes or trees, efficient and agile swimming robots, and lightweight flapping-wing flying robots. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-09

Project Summary This proposal centers on developing new deoxygenative functionalization of alcohols and carboxylic acids for accessing medicinally valuable chiral amines and fluorinated compounds through new sequential, stereo- controlled reactions. Alcohols and carboxylic acids are ubiquitous, structurally diverse, and stable feedstock chemicals, making the asymmetric deoxygenative diversification of their C–O bonds a highly attractive strategy for exploring new chemical space and discovering new therapeutics. While seminal approaches have demonstrated converting alcohols and carboxylic acids to pharmaceutically relevant chiral products, these methods often require multiple steps, undergo slow C–O bond activation, are limited in nucleophile scope, and lack stereo-control. To overcome these challenges, the Kim Group will employ a sulfonyl fluoride as a SuFEx (Sulfur Fluoride Exchange) clickable reagent for alcoholic C–O bond activation and a catalytically versatile base metal that will facilitate multiple C–O bond functionalizations of carboxylic acids. Our methods will allow one- step, rapid asymmetric C–O bond derivatization to construct C(sp3)–N/C bond formations between unfunctionalized starting materials and diverse nucleophiles with predictable stereochemical outcomes. Herein, we will develop the following Research Programs: (1) One-step, stereospecific and site-selective deoxygenative amination and perfluoroalkylation of alcohols via SuFEx, allowing direct installations of pharmaceutically relevant free amines, N-heterocycles, and perfluoroalkyl groups into alcohols, and (2) enantioselective deoxygenative diversification of carboxylic acids, providing a new platform for chiral amine synthesis via sequential geminal C– O bond functionalizations of a carboxyl group. Together, the successful realization of the goals described in these research programs will provide a new set of robust tools for rapid asymmetric derivatization of C–O bonds within widely available feedstock chemicals and complex molecules, which will expedite synthetic access to medicinally valuable products.

NSF Awards · FY 2025 · 2025-09

This National Science Foundation Innovations in Graduate Education (IGE) award to Clemson University will develop and evaluate an artificial intelligence (AI)-driven Virtual Teaching Assistant (ViTA) designed to support interdisciplinary graduate education in medical biophysics. Graduate students in this field face unique challenges due to the complexity of integrating concepts from physics, biology, chemistry, and engineering. These challenges are compounded by the lack of structured human teaching assistant support in many advanced courses. ViTA is an innovative solution that uses expert-validated generative AI to deliver personalized, context aware academic support within existing graduate curricula. By offering real-time feedback and interactive engagement, ViTA aims to improve learning outcomes, support a wide variety of student populations and contribute to institutional efforts to improve scalable graduate education. ViTA integrates fine-tuned generative AI models with retrieval-augmented generation pipelines to provide expert-validated, domain-specific feedback embedded directly into a learning management system. The project will use a mixed-methods research design to evaluate ViTA’s impact across cognitive, behavioral, and affective learning domains. Assessment strategies include validated pre/post tests, surveys of self-efficacy and cognitive load, reflective journals, and AI interaction analytics. Control and experimental cohorts will be drawn from students at the same academic level, including those taking medical biophysics courses as electives. Findings from this project will generate new knowledge on AI-supported interdisciplinary education, inform ethical pedagogical use of AI in graduate instruction, and provide a scalable framework that can be adapted across STEM disciplines. All tools, datasets, and protocols will be openly shared through public repositories and institutional partners to maximize replicability and long-term impact. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

This project researches the development of a comprehensive framework that ensures the principled use of artificial intelligence (AI) technologies in data-intensive education research. The framework will benefit society by enabling more trustworthy education research, fostering public confidence in AI applications, and ensuring that technological advances serve all students. The project provides insights on responsible AI practices in education research and supports education by creating publicly available training materials for researchers and others with varying technical backgrounds. The project aims to bridge the gap between high-level principles and practical implementation by developing a responsible and principled framework for data-intensive education research in the AI era. The framework targets three key stakeholder groups: data administrators who manage access to education data; researchers who use education data to perform analysis; and individuals who are deciding whether to participate in research studies. The project uses a mixed method study with all three stakeholders to understand their current practices, concerns, and decision-making processes regarding education data usage. Based on these insights, the project team will develop and deploy a novel web-based assessment tool that leverages state-of-the-art responsible AI techniques to detect potential risks within datasets, helping stakeholders make informed decisions about data sharing and usage. Additionally, the project team will create a toolkit that identifies bottlenecks from the community and translates complex AI risks and benefits into accessible formats, utilizing interactive visualizations to facilitate understanding among non-technical stakeholders. Finally, the team will develop comprehensive educational support materials, including video tutorials, interactive modules, and real-world case studies that demonstrate principled AI practices in education research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Research security mandates are being created and implemented across multiple sectors, and this creates a need for systematic and empirical insights into how these policies affect compliance and collaboration decisions of U.S. researchers. Traditional surveys fail when probing sensitive behaviors: participants doubt the provided anonymity guarantees and tend to underreport non-compliance. This project will create a gamified behavioral experiment to establish base rates for security compliance behaviors and formulate researchers’ compliance decision-making. This approach will overcome the prevalent reluctance among researchers to participate in “Research on Research Security.” The decision rules revealed via the behavioral data in this project will enhance theoretical models of human-policy interaction, offering transferable paradigms for behavioral studies in other sensitive domains. This project will establish methodological best practices and empirical benchmarks to advance the research security field. Findings of this project will help policymakers, funding agencies, and research institutions with evidence-based insights to craft adaptive security policies that preserve scientific openness while mitigating risk. This project will also train graduate students in the field of Research on Research Security. This project will create a novel approach that combines interviews with gamified scenario-based behavioral experiments to mitigate biases in self-reported data and to advance our understanding of compliance decision-making of U.S.-based researchers. The research will begin by creating and validating a taxonomy of security-relevant collaboration scenarios through semi-structured interviews with a broad group of U.S. researchers, particularly in the AI field. The taxonomy will uncover common patterns of international research collaboration, such as sharing code or data, and generate a library of authentic scenarios. Then, these scenarios will be used in gamified behavioral experiments to establish security compliance base rates in international collaborations. The game will capture realistic trade-offs without direct questioning, enabling us to derive rigorous base rates of security-related behaviors among researchers who might otherwise withhold accurate survey responses. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-09

Features are used to describe the characteristics of objects. For example, "age", "smoking or not", and "years of smoking" are features of a patient, which can be used to describe the patient's physical condition, and furthermore, to predict if she or he is likely to get lung cancer. A combination of features could be more helpful to the prediction, e.g., "age" minus "years of smoking" can be a new feature to indicate how early the patient starts smoking. This kind of feature combination is called feature generation. In the big data era, there exist enormous numbers of features, and it is not realistic to generate features manually by human experts. This project will build new technologies to automatically generate new features based on existing features, to better describe the objects, and to gain better prediction performance. Additionally, this project aims to substantially improve the traceability, affordability, and explainability during the generation process. The developed algorithms and tools are expected to be generalized and applicable to a broad range of scientific and engineering problems, not just in feature generation, but also in other domains such as data pre-processing, social analysis, intelligent transportation systems, healthcare, and the internet of things. This project identifies three research tasks: (i) A Reinforcement Learning (RL) based approach to realize traceability. Two RL agents are used to select appropriate features, and one RL agent is used to select the appropriate operation. The policy network will be decomposed into two sub-networks, i.e., representation network and value network. Different agents will share the value network to improve training convergence. (ii) A heuristic approach to realize affordability. Information theory-based utility scores will be designed to evaluate features and feature sets, and the heuristic selection strategy will be designed in the generation process. (iii) A Large Language Model (LLM) based approach to realize explainability. The tabular data will be serialized into natural language strings, and comprehensive prompts will be designed incorporating feature generation expertise and domain expertise. The LLM can generate features with explanations by fine-tuning it with prompts. Two strategies will be proposed to compress the prompt. The proposed research will provide novel perspectives and methodologies as to how to generate new features by advancing the understanding and designing new generation strategies. They go beyond conventional generation methodologies that are highly dependent on domain knowledge. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

The past four decades have seen a significant increase in wildfire frequency, magnitude, and resulting human and economic losses as driven by climate change and rapid population growth into the wildland-urban interface (WUI, where homes and infrastructure meet the wildland). Wildfire risk can be significantly reduced by three types of pre-event household hazard adjustments—mitigation, preparedness to stay, defend, and survive, and readiness to evacuate. However, the factors that influence the adoption of these hazard adjustments remain poorly understood. Existing wildfire evacuation models rarely consider fire spread dynamics; lack trilateral integration of people, fire hazard, and traffic components; and are based on limited social-behavioral data. To address these knowledge gaps, this project will integrate behavioral data with active learning and goal-setting techniques for increasing WUI residents’ adoption of pre-event hazard adjustments. In addition, social-behavioral data will be infused into transportation engineering models to create more accurate and actionable agent-based models (ABMs) for evacuation. To achieve these objectives, the researchers will collaborate with four WUI communities in three states to (1) identify factors influencing households’ pre-event hazard adjustment adoption and evacuation decision-making for wildfire hazard, and (2) integrate social-behavioral data into wildfire evacuation scenarios using ABMs to evaluate alternative evacuation strategies. Regional planners and emergency managers will be engaged to test and evaluate evacuation protocols and educational programs. This project will expand and strengthen the capability of the Protection Action Decision Model (PADM) to explain complex decision-making processes related to wildfire mitigation, stay/defend/survive preparedness, and evacuation readiness. Specifically, this project’s results will advance our knowledge in pre-event risk messaging about wildfire hazards and address the urgent need for incorporating multidimensional datasets in wildfire evacuation models. The study of four different WUI communities will allow assessment of the cross-population generalizability within and beyond the project’s study areas. The outcomes will lead to best practices for motivating households’ protective actions, assessments of community-informed evidence-based strategies for wildfire evacuation modeling, testing of alternative wildfire warning messaging strategies, and, ultimately, reduction in wildfire risk to residents and businesses. The diverse project team includes early-career scientists, students, and researchers from underrepresented groups. The project will build upon previous collaborations with community stakeholders to co-produce and share knowledge throughout the research process, ensuring that the work will promote data-driven policies and resource allocation. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Cells in the human body experience a variety of physical stresses, especially when they move and squeeze through tight spaces. Studies have shown that some cells become stronger after experiencing physical stress. Furthermore, some cells "remember" stressful experiences and change their biological responses as a result. These physical stresses change the physical appearance of the cells, but it is not understood if these physical changes cause biochemical changes inside the cell. The goal of this project is to engineer a specialized set of devices to understand how physical deformation of cells changes their ability to divide and move. These miniaturized devices simulate the tight spaces the cells squeeze through in the human body. These devices have the unique ability to study individual cells and the overall cell population. The cells can be observed over extended periods of time to study the "memory effects" of physical stress, which will improve understanding of how cells adapt and survive in the human body. This research can lead to new insights into how wounds heal, how cancer spreads through the body, and how stem cells move to where they are needed. This project is expected to have a broad impact on inspiring and preparing the next generation of scientists and engineers through research and activities that include middle school outreach, high school research opportunities, and international experiences for undergraduate students. Human cells constantly face various biophysical stresses that lead to increased proliferation and motility. Exposure to biophysical stressors can also result in a memory effect where the enhanced phenotype is retained for several days after the biophysical exposure. However, the relationship between cellular deformation and the resulting biochemical signaling has not been fully elucidated. To address this, a microfluidic deformation platform to mimic physiological cell squeezing will be engineered. This platform will be used to test the hypothesis that biochemical changes drive enhanced proliferation and motility that are governed by growth factor receptor expression and the stemness-dependent PI3K-AKT-mTOR signaling and other proliferation-related pathways. Objective 1 will study how biophysical constriction alters the phenotype of deformed cells that result in enhanced proliferation by pairing a deformation module with a trapping device. This will demonstrate that deformation enhances activation of proliferation-related pathways and that it will exhibit a memory effect in both single cells and the bulk population. Objective 2 will determine how biophysical constriction alters the motility of the cells by pairing the deformation module with a migration device to study one-dimensional single cell migration and the bulk two-dimensional migration. The signaling pathways that govern how deformation enhances cell motility and how that deformation results in a memory effect will be studied. Since cell deformation is vital in processes like wound healing, cancer metastasis, and stem cell homing, gaining a deeper insight into how it primes the cells for increased proliferation and motility is essential for future medical breakthroughs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-08

Understanding how fluids interact with elastic or porous materials, such as blood flowing through arteries or water filtering through soil, is crucial for addressing practical challenges in medicine, environmental science, and engineering. This type of modeling, known as fluid-structure interaction (FSI) or fluid-poroelastic structure interaction (FPSI), involves complex physics and mathematics because the fluid and the structure influence each other in strongly coupled ways. Traditional domain decomposition approaches solve the fluid and structure parts separately, then iteratively exchange information, but this process can be computationally expensive, especially for large systems. This research aims to develop new and efficient numerical methods that can solve these coupled problems more accurately and with reduced computational cost, enhancing simulation capabilities across diverse applications from hemodynamics to subsurface flow. This project is on a rigorous development and analysis of numerical schemes for solving FSI and FPSI problems, using a unified, monolithic framework with Lagrange multipliers to enforce interface conditions. The core of the work includes: (1) formulating and analyzing a monolithic system to establish well-posedness, stability, and finite element error estimates; (2) deriving a Schur complement equation to decouple the subdomain problems and enable efficient computation of the Lagrange multipliers; (3) applying projection-based reduced order modeling (ROM) to the Schur system, stabilized by supremizer enrichment, to reduce computational cost; and (4) extending the framework to a novel three-dimensional fluid–two-dimensional plate interaction model. These advancements aim to significantly enhance the computational efficiency and robustness of simulations for strongly coupled multiphysics systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-08

Project summary My research is focused on developing imaging tools and data analysis in single-particle tracking (SPT) and single molecule localization microscopy (SMLM). My long-term goal is to push their resolution limits to facilitate the study of protein interactions and organizations in cells at high spatiotemporal resolution. In the next five years, my lab will focus on two research directions: 1) developing high-resolution two-color SPT to study dimer formation and dissociation; 2) unifying data analysis for SMLM to extract maximum information and achieve expert-free analysis. SPT has been widely used in studying protein dynamics and interactions both in vitro and in live cells. However, the commonly used camera-based SPT methods are limited to low spatiotemporal resolution. Real-time SPT techniques, such as MINFLUX, offer significantly higher spatiotemporal resolution but are limited in their applications to two-color tracking. We will develop a two-color SPT microscope based on real-time SPT techniques and apply the developed system to study the interactions of epidermal growth factor receptor (EGFR) in live cells. EGFR, which plays an important role in regulating cell growth, motility and differentiation, is an attractive candidate for anticancer therapy. SMLM, as one of the major super-resolution microscopy techniques, can achieve a spatial resolution down to 2 nm and has been increasingly used in structural biology to bridge the gap between electron microscope and confocal microscope. Current SMLM data analysis pipeline consists of extracting the point spread function (PSF), localization using the obtained PSF model, drift correction and post-analysis such as clustering and particle averaging. This multi-step approach often complicates the data analysis procedure, as each step requires a certain level of expertise. Taking advantage of the large-data processing capability of deep neural network (DNN), we will develop a data-analysis framework based on DNN to achieve simultaneous localization and PSF extraction from SMLM data. This framework can be extended to any SMLM system and has the potential to extract all information from the data. I believe that the development of fast, high-resolution, two-color SPT techniques will significantly advance the study of protein interactions in cells. At the same time, a unified data-analysis framework for SMLM will render SMLM techniques more robust and accurate, thereby establishing them as a standard tool for structural biology.

NSF Awards · FY 2025 · 2025-08

This goal of this project is to turn two greenhouse gases — carbon dioxide (CO₂) and methane (CH₄) — into methanol, which is a useful chemical and a cleaner fuel. The process uses a plasma, which is a special state of matter made from superheated particles, with clean energy such as solar power. Instead of capturing and storing CO₂, this project takes it from the air near places where methane is released and turns it into methanol. This approach could reduce costs, make use of the U.S.'s natural gas supplies, and improve national security. It could also boost the economy by creating new markets for these gases. The project will train college students in science and technology and include hands-on activities to interest younger students in science and engineering. While conventional thermal dry methane reforming (DMR) transforms CO2 and CH4 into CO and H2, the project team has demonstrated that the use of perovskites and a non-thermal plasma route promotes the direct production of methanol. Successful completion of this project will advance the understanding of the plasma-catalyst interactions at the nanoscale responsible for the oxygenate yield. Another intellectual contribution will be the development of plasma-activated/plasma-enhancing perovskites and dual-function metal-perovskites for CO2 capture and conversion, which can be applied across a wide range of approaches. Additionally, a comparative study of the thermal vs. plasma route will provide insights into the isolated effects of the fields and plasma-charged particles. The objectives for this project are to fundamentally understand: (1) the interactions between plasma-originated species and perovskites under different plasma scenarios and their implications for reactivity, (2) the mechanistic impact of plasma properties and composition and individual species during CO2 conversion, (3) the material response (charging, polarization, electric field) to plasma and surface properties of perovskites that enable the CO2 catalytic conversion to methanol, (4) the link between surface properties and plasma conditions and the boundaries between thermal and plasma effects, and (5) the link between material composition and CO2 binding and reactivity. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2025 · 2025-08

PROJECT ABSTRACT While modern medicine has seen many advances that have improved human health over the last century, most medications available to the public do not act on the cause of disease, mainly focusing on alleviating symptoms associated with downstream effects. The overarching goal of my research program is to develop nanoparticles with pathologically responsive chemistries to enable the targeted delivery of disease-modifying biologic agents. One such focus is on the use of RNA-based therapies. Compared to traditional methods of altering gene or protein expression that involve editing or modification, RNA-based therapies offer several advantages, including a lower risk of off-target genetic mutations, temporary and controlled therapeutic gene expression, and a shorter production time, which facilitates rapid responses to various emerging health challenges. However, the translatability of RNA-based therapeutics relies on some sort of delivery vehicle or modification to assist in biologic transport, like a nanoparticle. Although delivery of nucleic acids in lipid nanoparticles (LNPs) has been successful, the effect is transient due to the transient nature and RNAse degradability of nucleic acids, which means multiple injections are often required to receive a lasting protective effect and therapeutic effects are minimized. Furthermore, LNPs notoriously traffic payloads to the liver, which does not enable their use in alternative tissue targets. In this proposal, we lay out a high-risk, high-reward approach to 1. Increase the stability of RNA in vivo, 2. Control the timing of RNA release and protein translation, and 3. Enable the delivery of RNA- based therapeutics into the brain, therefore addressing the current translational concerns and extending RNA therapy to tissues that are difficult to reach using polymer-based nanoparticles, called polymersomes (Figure 1). Furthermore, although our initial focus is on mRNA, our approach could be widely applied to other negatively charged RNAs. Stimuli-responsive coacervated mRNA will be encapsulated in polymersomes and delivered across the blood-brain barrier, enabling RNA translation on demand in brain tissue for the first time.

NSF Awards · FY 2025 · 2025-07

Data reduction holds paramount importance in scientific endeavors and various data-intensive domains. This necessity is compounded by the unprecedented surge in data volume propelled by advancements in facilities and scientific research. Data compression stands as a prevalent method for data reduction. But prevailing solutions are hindered by a fundamental constraint: the requirement for decompression prior to processing. This introduces three primary challenges: (i) heightened strain on storage and memory resources, (ii) potentially lengthy decompression times for sizable datasets leading to significant workflow delays, and (iii) sometimes loss of accuracy when applications are compelled to operate on partial data views for space shortage. Aiming to develop a set of novel programming and runtime techniques, this project will create Efficient Processing without Decompression (EPOD), a novel approach to data reduction by lossless data compression while maintaining direct processability (without decompression) in the compressed state. The success of the project is poised to yield significant impacts, potentially reducing data sizes in numerous domains by one or two orders of magnitude without quality loss, while concurrently expediting data processing by multifold. The technique will help accelerate scientific research as well as improve the efficiency and productivity in various data-intensive domains. To develop the idea of EPOD into a new paradigm and solution for data reduction, this project includes five research activities: (i) developing the basic EPOD method and expanding its data coverage to floating-point datasets and so on; (ii) creating multi-level support of EPOD operations (i.e., data accesses or manipulations working directly on the EPOD compressed format) in forms of an EPOD library and a scalable large language model-based EPOD synthesizer; (iii) enabling continuous compression for streaming to make EPOD seamlessly integratable in streaming workflows of scientific facilities; (iv) developing advanced optimization of EPOD to maximize the efficiency and scalability; (v) integrating with existing data analytics ecosystem and complementary techniques and applications, and demonstrating the impact on a broad range of scientific domains and applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NIH Research Projects · FY 2026 · 2025-07

PROJECT ABSTRACT The blood-nerve barrier (BNB) is a highly selective boundary that often prevents the non-invasive delivery of drugs to nerves, which is why nerve injuries and diseases remain difficult to treat1. The BNB highly limits the non-invasive treatment of peripheral neuropathies, meaning that there are few treatments available depending on the cause, despite severe cases that can result in debilitating nerve alterations. Even with traumatic peripheral nerve injuries where regeneration spontaneously occurs, functional recovery is often very poor since regeneration is abysmally slow. Current gold standard for addressing these traumatic nerve injuries involves surgical interventions. However, even with these surgical interventions, most patients do not obtain functional recovery if regeneration distances exceed a few centimeters. Here, we propose the design and implementation of a local and systemically administered nerve-targeting polymersome platform that can penetrate peripheral nerves and lead to functional recovery after peripheral nerve crush injury. We will investigate the effect of adding two targeting ligands, Apolipoprotein E (ApoE) and Rabies Virus Glycoprotein peptide RVG-9R, to the polymersome surface on BNB penetration. Preliminary data suggest that RVG-9R enables delivery through the potentially injured blood-nerve barrier and ApoE is assists in the retaining polymersomes in the nerve. We will elucidate the mechanisms associated with these ligand-based approaches by studying alterations in receptor expression on the BNB and neural cells. We will study the pharmacokinetics of ligated particles after sciatic nerve injury, correlating changing behavior with receptor expression. We will then investigate the nerve regenerative effects of two different biologic drugs developed by co-I Jeff Twiss, a G3BP1 peptide and siRNA targeting prenyl-Cdc42 mRNA, delivered through polymersomes. This proposed work represents a novel, potentially paradigm shifting approach to selectively deliver therapeutics to the peripheral nervous system through non-invasive systemic injections, with the goal of enabling functional recovery post nerve injury. However, the modular nature of the polymersome system and the ability to substitute various therapeutic payloads enables the extensions of findings to acquired and inherited nerve diseases.

NSF Awards · FY 2025 · 2025-07

With the support of the Macromolecular, Supramolecular, and Nanochemistry Program in the Division of Chemistry, Dr. Rhett C. Smith of Clemson University will investigate a new method for converting agricultural and plastic waste into useful materials through a process called thiocracking, where sulfur and heat are used to break down organic molecules. This research will address challenges in waste disposal, including plastics, agricultural residues, and mixed-material waste, while using excess sulfur, a byproduct of fossil fuel refining. Thiocracking could transform these waste streams into recyclable materials for construction and infrastructure, reducing landfill waste, and benefiting the U.S. economy. The project will also support education by training graduate and undergraduate students, collaborating with primarily undergraduate institutions, and sharing findings with the public through social media and open-access publications where possible. The project will specifically explore a two-step thiocracking-desulfurization sequence to break down complex waste streams and recover valuable organic products. In the first step, thiocracking will be used to induce depolymerization and conversion into polymeric composites. These composites will then be assessed for mechanical properties and recyclability, with a focus on their potential applications as structural materials. The second step will employ catalytic hydrodesulfurization to remove sulfur and recover hydrocarbons. This process will be optimized for catalyst performance, reaction conditions, and product distribution. Mechanistic studies will provide insight into sulfur-carbon bond formation / cleavage, advancing the fundamental understanding of thiocracking chemistry. If successful, this research would lead to an innovative, energy-efficient, and scalable strategy for repurposing waste plastics and biomass, offering a transformative solution to waste management through chemistry and materials science. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Awards · FY 2025 · 2025-07

The project aims to conduct a systematic study of quantum harmonic analysis, an emerging new area that extends the core ideas of classical harmonic analysis into the quantum setting. While classical harmonic analysis is concerned with representing functions or signals as combinations of basic waves (like sine and cosine functions), quantum harmonic analysis applies these methods to quantum states and observables within the framework of quantum mechanics. This field plays a foundational role in quantum information science, providing the mathematical tools needed to describe, analyze, and manipulate quantum systems. By developing a deeper theoretical understanding, the project will help lay the groundwork for future breakthroughs in quantum science and technology, including quantum computing, quantum cryptography, and quantum communication. The project also provides research training opportunities for graduate students. The project proposes new methods from topological dynamics and harmonic analysis to address numerous open problems in quantum harmonic analysis. In addition to tackling foundational questions, the project introduces several new research directions aimed at advancing this young and rapidly evolving area. The scope of the project includes a broad spectrum of problems with varying levels of difficulty, some specifically designed to be accessible to graduate students, thereby fostering involvement from early-career researchers. Moreover, the project provides a comprehensive framework to help organize, systematize, and clarify the recent surge of activity in quantum harmonic analysis. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.