Purdue University

NSF Awards · FY 2026 · 2026-10

Pre-trained AI models shared through open online repositories are becoming essential infrastructure for research, industry, and government. But this growing reliance also creates an important cybersecurity concern: just as traditional software can be attacked to include viruses or access backdoors, AI models can also be tampered with. This can lead to security breaches and errors in systems that rely on these pre-trained models. This project will develop methods and tools to help users verify whether a pre-trained AI model is trustworthy before it is incorporated into scientific workflows, operational systems, or other important computing environments. By improving the security of this emerging AI infrastructure, the project will help strengthen the U.S. research enterprise, support economic competitiveness, and improve the resilience of AI-enabled systems. The project will also advance education and workforce development by training students, providing research opportunities, and fostering collaboration among universities, industry, and other stakeholders. This project develops a novel approach to address three major security challenges in the machine learning (ML) model supply chain. The research integrates software engineering principles with machine learning techniques to systematically mitigate vulnerabilities during model selection, loading, and management. First, the team of researchers will tackle model spoofing, where adversaries upload malicious models using deceptive names. The project relies on novel anomaly detection schemes for naming conventions and architectural signatures to identify these threats. Second, the investigators will secure the model deserialization process. Because frameworks often use formats vulnerable to arbitrary code execution, the research will develop automated, least privilege deserialization mechanisms and define safe subsets for model loading runtimes. Third, the project will establish robust model lineage tracking to manage the risks of reusing models. The team will create a lineage graph data structure that combines static and dynamic analysis to trace model evolution and detect illicit modifications. By integrating these methods, the project provides a comprehensive defense system that enhances trust, integrity, and oversight in the open source model ecosystem. 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 2026 · 2026-09

Software development is rapidly changing, with developers increasingly delegating programming tasks to code generation tools powered by Large Language Models (LLMs). These systems often produce code that appears correct, but that actually contains subtle defects and inconsistencies with the developer’s intent, compromising the reliability and security of the generated program. As the role of software engineers moves away from writing code directly to inspecting generated programs, it is vital that they have trustworthy tools that help them rigorously examine and understand program behavior. Recent advances in formal methods have demonstrated that it is now possible to verify that complex real-world programs like compilers and operating systems behave exactly as intended. However, these tools are designed to confirm that a well-understood and correct program always does the right thing and are ill-suited to helping a developer explore and reason about a partially understood program that may contain flaws. The project’s novelties are the development of new, principled techniques for rigorously reasoning about the behaviors of potentially incorrect programs, and the creation of a trustworthy, general-purpose framework that helps developers effectively explore and understand such programs. The project’s impacts are increased trustworthiness and accountability in AI-assisted software development, supporting the responsible integration of AI code generation tools into larger software toolchains, and helping developers confidently deploy AI-generated software in safety- and security-critical domains. The major technical innovation of this work is a novel kind of type refinement that soundly captures the existence of certain paths that lead to errors, while soundly pruning those that are irrelevant to the property of interest. This type abstraction is capable of reasoning about the existence of “buggy” or undesirable executions and can soundly establish that a program is incorrect. The resulting reasoning framework leverages automated theorem provers to perform this analysis automatically and explores new modes of interacting with these solvers in the face of partial knowledge about a system. The educational goals of the project are to equip learners at all levels with principled reasoning techniques for understanding computer programs, emphasizing the foundations of automated and symbolic reasoning and their application to modern and AI-assisted software development. 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 2026 · 2026-09

Personal care products may release chemicals into indoor air during their normal use. These chemicals may stick to surfaces, form small airborne particles, and move through ventilation systems to outdoor air. These processes play affect indoor air quality and its link to outdoor air. However, the basic processes that control how these chemicals are released, changed, and moved inside buildings are not well known. This CAREER project will examine how product usage, heat, and building operations affect chemical behavior indoors. The research will use laboratory tests and full-scale experiments in a residential research test house representative of indoor conditions. The work will also support AI-based models of chemical movement and ventilation control in buildings to improve indoor air quality. It will advance STEM education through hands-on engineering labs, service-learning design projects, and public demonstrations using a mobile test house. The CAREER project will quantify emission kinetics, multiphase transformations, gas–surface interactions, gas–particle partitioning, and ventilation-driven transport of volatile and semivolatile compounds associated with personal care product use in residential buildings. Controlled emission cell experiments will resolve composition- and temperature-dependent release rates across representative product classes, while full-scale experiments in a residential research test house will quantify airborne persistence, surface sorption and re-emission, nanoparticle nucleation and growth, and ventilation-mediated removal under variable building operating conditions. High-resolution, real-time chemical measurements using online mass spectrometry and advanced aerosol instrumentation will generate volatility- and composition-resolved gas- and particle-phase datasets. Where appropriate, molecular descriptors derived from quantum chemistry methods such as density functional theory will support interpretation of measured chemical reactivity and phase-partitioning behavior and inform AI-assisted fate and transport modeling. Mechanistic modeling frameworks integrating material balance analysis, volatility basis set representations, and indoor-to-outdoor flux quantification will link product characteristics, indoor environmental conditions, and building operation to chemical fate and transport metrics. These activities will produce a comprehensive, volatility-resolved emissions characterization framework for personal care products and establish scalable process models to advance AI-enabled indoor air quality simulation and building system design. Educational activities will help develop a U.S. engineering workforce with expertise in building mechanical systems, indoor environmental monitoring, and AI-driven building design. Undergraduates will engage in service-learning air quality design projects through Purdue’s Engineering Projects in Community Service (EPICS) program, will apply research-grade data in hands-on laboratory modules embedded in civil and construction engineering courses, and will participate in full-scale experimentation using the Purdue zero Energy Design Guidance for Engineers (zEDGE) Test House. Intergenerational learning through Purdue Grandparents University and mobile outreach activities at libraries using zEDGE will strengthen students’ communication and systems-thinking skills while translating engineering concepts to the public. 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 2026 · 2026-08

NONTECHNICAL SUMMARY A central challenge for quantum materials is understanding how large numbers of electrons interact and organize themselves into collective quantum states. These interactions often lead to emergent behavior, where the system exhibits new properties that are more than those of the sum of its parts. Well-known examples include superconductivity, which enables electric current to flow without resistance, and magnetism, which enables materials to generate magnetic fields. These phenomena have led to mature applications in technologies such as medical imaging and magnetic storage. Even more exotic is the phenomenon of charge fractionalization, where the material behaves as if its electrons have split up and carry a fraction of the electron charge. Such effects could potentially be harnessed to enable noise-resilient quantum information processing. Recent advances in experimental techniques have made it possible to fabricate and measure mesoscopic-scale quantum devices that serve as simplified yet powerful platforms for studying these interactions. This project seeks to deepen our understanding of theoretical models and to bridge the gap between theory and experiment by making testable predictions using advanced analytical and numerical techniques. By predicting signatures of complex electron interactions in relatively simple quantum devices, the research will propose new approaches to create and probe electronic states that go beyond conventional theories. The project also emphasizes education and workforce development. Graduate and undergraduate students will receive training in advanced theoretical and computational methods and will participate in research at the frontiers of quantum science. Outreach and classroom activities will include hands-on demonstrations of quantum entanglement and the development of a modern course on superconductivity, both available to the general public. These efforts aim to broaden public understanding of quantum physics and prepare a highly skilled quantum workforce, strengthening U.S. leadership in quantum information science. TECHNICAL SUMMARY This project investigates strongly correlated quantum systems through the theoretical study of quantum impurity models where both the impurity and its environment can be topologically nontrivial. These systems are perhaps the simplest examples that exhibit rich physics including non-Fermi liquid behavior, emergent anyonic excitations, and unconventional symmetry structures beyond SU(2). Recent advances in mesoscopic device fabrication and materials development have made it possible to probe these paradigmatic models experimentally, while theoretical progress has revealed new opportunities to explore the interplay of topology, symmetry, and electron correlations. The project aims to bridge the gap between theory and experiment by making testable predictions using analytical and numerical techniques. The research will focus on elucidating electron correlations and topology through quantum transport in mesoscopic devices, probing topological boundary excitations using quantum impurities, and developing new tools to study correlated multi-impurity states. The work combines quantum many-body and linear response theory, conformal field theory techniques, as well as numerical methods such as density matrix renormalization group. A key goal is to clarify the properties of emergent anyons in quantum impurity models, including Kondo anyons that arise in gapless systems and exhibit nontrivial impurity entropy, and to determine how their behavior compared to anyons in gapped topologically ordered phases. The project also emphasizes education and workforce development. Graduate and undergraduate students will receive training in advanced theoretical and computational methods and will participate in research at the frontiers of quantum science. Outreach and classroom activities will include hands-on demonstrations of quantum entanglement and the development of a modern course on superconductivity, both available to the general public. These efforts aim to broaden public understanding of quantum physics and prepare a highly skilled quantum workforce, strengthening U.S. leadership in quantum information 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 2026 · 2026-08

Modern computer vision has advanced rapidly in image understanding, editing, and generation. However, strong performance alone is insufficient for real-world deployment. Practical systems must also be safe, accountable, interpretable, and maintainable, especially as AI becomes deeply embedded in everyday life. Today’s vision systems are typically built on large, fully parametric models that encode vast amounts of training data in complex, non-intuitive ways. This makes it difficult to diagnose undesirable outputs, adapt models to shifting data distributions, or trace specific behaviors back to the data that influenced them. This project addresses these limitations by developing a new class of data-centric models, where model behavior can be more directly interpreted, attributed, and updated through explicit connections to training data. The goal is to enable more controllable and maintainable AI systems. Prior work has explored post-hoc approaches for data-centric capabilities; these methods are applied after a model is trained. While useful, these techniques do not address the underlying issue of fully parametric model design. In contrast, non-parametric methods such as k-nearest neighbors naturally provide strong data-centric properties, since predictions can be directly linked to training examples, though they often fall short in performance compared to modern deep models. This project proposes a semi-parametric paradigm that combines the strengths of both approaches: the performance of parametric models with the interpretability and controllability of non-parametric methods. The research is organized around four thrusts: developing semi-parametric architectures as a core framework and applying them to problems of unlearning, data attribution, and model customization. These models explicitly incorporate training data at inference time, enabling predictions to be traced back to specific examples. The project will further investigate how to ensure models effectively utilize this data at test-time, through architectural design (e.g., equivariance) and improved training strategies. 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 2026 · 2026-07

From smart devices to data centers, future artificial intelligence (AI) will need stronger capabilities for reasoning, logical thinking, and multi-step problem solving in dynamic real-world environments. Neuro-symbolic AI, which combines the strengths of neural networks and symbolic reasoning, is a promising direction for giving AI systems these capabilities. Yet such workloads remain difficult to run efficiently on today’s computing platforms because they place stringent demands on hardware performance, energy efficiency, programmability, and scalability. This project addresses that gap by developing new computing foundations for neuro-symbolic AI through cross-stack co-design, specialized memory technologies, and advanced three-dimensional integration. The goal is to create versatile, efficient, and scalable computing chips and systems that support more capable, real-time cognitive AI. In parallel, the project will develop new course materials and hands-on learning experiences in neuro-symbolic AI and semiconductors for students and K-12 educators, enhancing participation and literacy while helping prepare a future semiconductor workforce. Together, these integrated research and education activities will advance the computing foundations needed for future AI systems that can reason, respond, and assist more effectively across many real-world domains. The project develops versatile, efficient, and scalable neuro-symbolic computing platforms on three-dimensional integrated circuits and systems. The research is organized around four interwoven thrusts. These include (1) establishing a co-design framework that bridges neuro-symbolic models, memory-centric architectures, and system-technology co-optimization across silicon CMOS, emerging devices, and 3D integration schemes; (2) building efficient yet programmable neuro-symbolic accelerator chips that exploit heterogeneous silicon and beyond-silicon compute-in-memory (CIM) fabrics together with a CIM-native, neuro-symbolic instruction set architecture; (3) developing tailored 3D integrated systems that combine new reconfigurable memory primitives and 3D stacking schemes enabled by CMOS-compatible oxide-semiconductor logic and ferroelectric transistors; and (4) creating a neuro-symbolic chiplet macro compiler that generates modular, silicon-calibrated hardware macros to enable a closed-loop workflow for continued algorithm-hardware co-design. Collectively, these efforts will advance the performance, efficiency, scalability, and programmability frontiers of neuro-symbolic computing 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.

NSF Awards · FY 2026 · 2026-07

This project promotes the development of new methods that make artificial intelligence systems more reliable, more data efficient, and easier to correct. Modern systems for language, images, and scientific data often require enormous training sets and computing resources, can become unstable during training, and are difficult to update when information must be removed for privacy or safety reasons. These limitations can hinder scientific discovery and make advanced computing less accessible. This project addresses these challenges by learning how the structure of data shapes the behavior of modern learning systems, with the goal of reducing computational cost, improving reliability, and supporting safer curation of learnt models. The project will also strengthen the future computing workforce through undergraduate and graduate research training, course-based projects, open software and educational materials, and hands-on outreach for school students and teachers on data, algorithms, and responsible artificial intelligence. The research studies how individual training examples shape the local geometry of the loss function in modern machine learning. It has three connected aims. First, it will characterize and improve optimization stability in deep neural networks, including modern predictive and generative models, by developing diagnostics and training methods based on curvature alignment across data. Second, it will design small data summaries and synthetic training sets that preserve the structure of the full learning problem, thereby reducing data and computational cost while maintaining performance. Third, it will develop efficient methods for removing the influence of selected training examples with minimal damage to the rest of the model. The project will evaluate these ideas on image, language, continual learning, and modern text and image generation benchmarks, and will release benchmarks and instructional modules to support reproducible research and 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 2026 · 2026-07

The colossal scale of Machine Learning (ML) systems today means that only powerful players with sufficient computing resources are able to participate in large-scale ML development. As a result, ML pipelines tend to lack transparency or auditing mechanisms. Reliance on a small number of service providers also jeopardizes availability and reliability. Alternatively, ML development can be distributed to a network of volunteer organizations and individuals, mitigating dependency on central suppliers and motivating users to donate their restricted or private data through transparent use of artifacts for public good. However, a distributed setting also opens up a large attack surface from malicious actors who can tamper with any step of the process. This project addresses this challenge by creating tools for an open, secure, and practical distributed ML development paradigm. The project’s novel contributions are centered around verification mechanisms for distributed and heterogeneous ML pipelines with private data. More broadly, this project helps stakeholder communities and individuals take part in large-scale ML development without compromising their privacy, contributing to the advancement of Artificial Intelligence (AI) technologies that benefit society. This project also integrates the proposed research into educational activities to train a workforce knowledgeable in capabilities and vulnerabilities of AI tools, as well as outreach initiatives to engage stakeholder communities and industry practitioners with research. The project is divided into three main tasks. The first task proposes verification techniques for distributed data pipelines, allowing data holders to contribute sensitive data with privacy guarantees while attesting to the legitimacy of their submissions. The second task studies proof-of-learning through reproducing computational steps. The research first develops analytical and empirical models for computation output error due to factors such as runtime optimizations and hardware non-determinism. Then, the research plan investigates attacker capabilities to compromise security and determines a secure error margin in practice with minimal impact on performance. The third task develops frameworks for privacy-preserving preference data collection from intended users for fine-tuning and alignment training. Additionally, this task explores optimization techniques to further reduce the computation and communication costs of the proposed solutions. 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 2026 · 2026-07

A Programming Languages Mentoring Workshop (PLMW) is organized as part of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), the premier forum in the field of programming languages and programming systems research, covering the areas of design, implementation, theory, applications, and performance. Several of these areas are directly connected to administration priorities in AI and Quantum. Several PLDI papers directly address these priority areas. PLDI 2026 will be held in Boulder, Colorado. The impact of the award relates to providing opportunities for students to receive mentoring from leading researchers and building the next generation of researchers and knowledgeable practitioners in programming languages and systems. A proper understanding of such topics is crucial to the implementation of tools and toolchains that use artificial intelligence (AI) and machine learning components. The award's broader significance and importance include building international community, lasting professional connections to create and implement tools, and enhancing education of students. By supporting the students, the school thus imparts training to the next generation of researchers in design and implementation of programming languages in both industry and academia, as well as to future application developers. 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 · 2026-06

Project Summary Voice and communication deficits are common in Parkinson’s disease. They occur early in the disease process, are progressive in nature, and significantly impact the quality of life of individuals by increasing stress, social isolation, and family/caregiver burden. However, despite their impact, vocalization deficits are rarely the focus of preclinical studies and are unimproved by current medical treatments that focus on the nigrostriatal dopaminergic system. Our failure in treatment development exists because we do not currently know the underlying mechanisms responsible for the complex vocalization deficits that occur in Parkinson’s disease. This gap in knowledge highlights the need for research that addresses vocalization deficits with an approach that allows for specific manipulation of different parts of the neurophysiological pathways involved in vocalization. Therefore, the goal of this project is to use a preclinical mouse model that allows for region-specific induction of pathology within discrete areas of the nigrostriatal and cranial sensorimotor system. Using this model, we will determine the specific vocalization deficits that result when pathology is present in the nigrostriatal system versus when pathology is present in the cranial sensorimotor system. We will also determine the modulatory effect of vocal training on region-specific vocalization deficits and underlying pathology. Results will provide a critical and necessary foundational understanding of the underlying mechanisms that lead to early voice and communication deficits in Parkinson’s disease. In addition, results of this proposal will aid in the development of no vel approaches to diagnose and treat voice and communication deficits related to Parkinson’s disease.

NSF Awards · FY 2026 · 2026-06

This grant provides travel support to approximately 40 graduate students and postdoctoral fellows to present their research at the 2026 Annual Technical Meeting of the Society of Engineering Science (SES2026), which will be held at Purdue University in West Lafayette, Indiana, 11-14 October 2026. The SES Annual Meeting is a premier international forum in mechanics, bringing together researchers across engineering, physical sciences, mathematics, quantum mechanics, AI/ML, and related disciplines. Through focused symposia and poster sessions, the conference promotes the dissemination of cutting-edge research and fosters interdisciplinary dialogue within the mechanics and materials community. The primary goal of this grant is to enable outstanding graduate students and postdoctoral scholars to present their work to a broad audience of leading experts and peers, thereby enhancing the visibility and impact of their research. Participation in SES2026 will provide these early-career researchers with valuable opportunities to receive feedback, generate new ideas, and establish professional connections that support long-term career development. By facilitating strong participation from trainees across a wide range of institutions, this project advances NSF’s goal of building, retaining, and recruiting the next generation workforce in mechanics of materials and structures including area such as quantum mechanics, AI/ML etc., strengthening both the research community and the nation’s scientific enterprise. 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 2026 · 2026-06

Modern computing systems increasingly incorporate learned components using techniques from machine learning and artificial intelligence. Engineering practice favors reuse over building from scratch. However, while for conventional software we know much about the re-use and adaptation of components, the correspondence for pre-trained models is an emerging and evolving concern. Engineers must decide which models to trust, how to adapt them, and how to document their behavior, often without shared standards or guidance. The project’s novelties are its systematic investigation of how model reuse resembles, and differs from, traditional software reuse, and the creation of practical methods that make these differences manageable. The project's broader significance and importance are reflected in a toolkit that enables more efficient engineering practices, lowering the costs of developing intelligent computing systems. The project also produces substantial educational materials to support K-12, undergraduate, and graduate students, as well as practicing professionals. Its result is improved United States economic competitiveness, greater academia-industry partnerships, and a deeper pipeline of engineers with AI skills for opportunities in industry, academia, and government. The project applies methods from human factors and software systems engineering to study how practitioners discover, evaluate, adapt, and maintain pre-trained models. It identifies best practices, constructs taxonomies of engineering behaviors, and develops novel tools to accelerate software engineering work. The resulting knowledge covers the full re-use cycle, including (1) techniques to facilitate the identification of pre-trained models; (2) techniques to support the evaluation and selection of such models; (3) a novel, ecosystem-spanning dataset of models for further analysis; and (4) a grounded-theoretic advance on software engineering theory that contrasts the reuse of statistical, data-dependent learned models with conventional software --- altering assumptions about modularity, specification, and verification. Together, the project provides a foundation for principled, efficient, and trustworthy reuse of artificial intelligence components in modern computing 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.

NSF Awards · FY 2026 · 2026-06

Magnetic storms have caused major damage to communications and energy infrastructure in the past. Such storms have the potential to cause trillions of dollars of damage in the United States. Society is more vulnerable to these risks in areas where the Earth’s magnetic field is weak. Continent-scale weak regions have grown over the past century. Studying ancient magnetic fields will help society understand how magnetic storm risks change and evolve. This has been very challenging due to the time-resolution of datasets. This project determines whether coral skeletons contain records of the magnetic field with greater resolution. Corals have the potential to yield yearly records spanning tens of thousands of years. This project aims to unlock coral magnetic records by investigating how corals become magnetized in modern reefs. Extracting magnetic records from corals could provide a new proxy for the pre-historic magnetic field. Constraining past variations of the magnetic field could help predict its future behavior, and its potential impact to society. This project investigates whether corals preserve paleomagnetic records with sufficient temporal resolution to reconstruct geomagnetic field behavior on human-relevant timescales. Preliminary results demonstrate that some modern corals carry stable magnetic remanences consistent with the ambient geomagnetic field at the time of growth. The project will produce detailed paleomagnetic, rock-magnetic, and microscopic analyses of modern corals to characterize the magnetic mineral assemblages, determine the stability and origin of remanent magnetization, and assess how reliably corals record the Earth’s magnetic field during growth. These data will be used to evaluate the feasibility of extending coral paleomagnetic records into the fossil archive. Unlocking coral paleomagnetism will provide high-resolution geomagnetic time series advancing fundamental understanding of Earth’s magnetic field. Improved knowledge of geomagnetic variability will enhance societal preparedness for geomagnetic storms that threaten critical infrastructure and public health. 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 2026 · 2026-06

This NSF Engineering Research Initiation (ERI) project aims to make the nation's hydropower fleet, a leading source of U.S. renewable electricity, more reliable under varying uncertainties in climate conditions. This project will advance a new paradigm for reservoirs operations by replacing rules built from historical records with artificial intelligence (AI) tools that adapt to droughts, floods, and other unprecedented conditions. This will be achieved by combining two complementary advances including a generative model that produces realistic, physically consistent climate scenarios spanning a wide range of futures, and an adaptive learning method that uses those scenarios to extract reservoir-operating policies that remain effective when the climate shifts. The intellectual merit of the project includes new methodology that reframes climate change as a learning problem requiring policies that transfer across conditions, rather than retraining for each new regime, advancing the science of reliable AI for critical energy infrastructure. The broader impacts of the project strengthen national energy resilience, workforce development, and public engagement in STEM through multiple coordinated efforts: publicly released datasets, software, and a hydropower visualization platform that lower barriers to climate-adaptive energy research nationwide; a new module in a renewable-energy course; equal research opportunities for graduate and undergraduate engineering students; and a student-led club that hosts laboratory visits and expanding STEM education outreach activities for K-12 students. The technical work is organized around two integrated thrusts. Thrust 1 develops a multi-head conditional generative adversarial network (GAN) that synthesizes multivariate hydroclimatic time series, including precipitation, temperature, snow water equivalent, inflow, and electricity demand across drought, flood, and stochastic-rainfall regimes, while preserving the physical relationships between variables that conventional statistical downscaling tends to lose. Thrust 2 develops a physics-informed deep reinforcement-learning (RL) framework for cascaded multi-reservoir operation that remains effective as hydrological conditions evolve under climate changes. The framework learns a shared state representation that captures features common to various climate conditions and regimes, so that a single operating policy can be deployed across climates and across reservoirs without retraining the full model. Reservoir mass-balance dynamics, storage and discharge limits, and downstream demand reliability are embedded directly in the simulator and reward function, ensuring that all operating decisions respect physical constraints. The framework will be evaluated on publicly available U.S. federal reservoir operating records coupled with the Catchment Attributes and Meteorology for Large-sample Studies (CAMELS-US) basin dataset, with head-to-head benchmarks against established climate-scenario generation and reinforcement-learning approaches. 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 · 2026-06

PROJECT SUMMARY Invasive fungal diseases are a growing global health problem, currently estimated to contribute to the death of 3.8 million people each year. Primarily affecting the immune-compromised and those with chronic underlying health problems, fungal diseases are challenging to treat and exhibit very high mortality rates. There are few effective treatment options and pathogens are developing resistance to our limited antifungal drugs. Therefore, development of new antifungal agents directed towards novel targets with unique modes of action is a high priority. The overall objective of the proposed project is to develop novel antifungal lead compounds directed towards a recently validated new drug target, the Cdc14 protein phosphatase. Cdc14 is required for virulence of a variety of fungal species, including the most common human pathogen, Candida albicans. It’s structure and enzyme specificity are highly conserved across the fungal kingdom, and are unique among phosphatases. The ability to design chemical compounds that selectively inhibit Cdc14 phosphatases was recently demonstrated by this group, including an inhibitor that exhibits biological activity in C. albicans cells and can reduce virulence in a mouse model of invasive candidiasis. The proposed project will build on these findings to further optimize Cdc14 inhibitor compounds to have enhanced affinity, selectivity, bioavailability, and drug-like properties and to develop and use suitable models for pre-clinical testing. In the initial R21 phase, three independent Aims will be pursued. In Aim 1, the initial Cdc14 inhibitor will undergo lead optimization and SAR analysis based on structural modeling and recent success with the human orthologs to improve affinity and selectivity. This aim will involve organic synthesis coupled with in vitro enzyme assays. The most-improved compounds will be screened for cellular toxicity and improved in vivo efficacy against C. albicans. In Aim 2, the in vivo mechanism of action of the compounds in C. albicans cells will be characterized to test if Cdc14 is the relevant target responsible for the observed biological activity. In Aim 3, the importance of Cdc14 for virulence of two other fungal pathogens, Candida auris and Candida glabrata, will be tested using mouse infection models to begin addressing the potential breadth of action of Cdc14 inhibitors against diverse fungal diseases. In the subsequent R33 phase, the following aims will be pursued. In Aim 1, the optimization and SAR data from the R21 phase will be used to design and synthesize additional generations of Cdc14 inhibitor derivatives intended to maximize affinity, specificity and drug-like properties. In Aim 2, affinity and selectivity of the new compounds will be compared in biochemical assays, and mouse toxicity and pharmacokinetics measured. Finally, in Aim 3, efficacy of the most promising compounds will be assessed in cultured cells and mouse models of invasive candidiasis caused by C. albicans, C. auris, and C. glabrata. The R33 work will involve iterative rounds of optimization guided by output from the biochemical and in vivo assays, eventually resulting in lead compounds with promising characteristics for full pre-clinical and future clinical testing.

NSF Awards · FY 2026 · 2026-06

This project supports U.S.-based researchers to participate in the five-day conference Heat Kernels and Stochastic Analysis, to be held at Aarhus University in Denmark from June 29 to July 3, 2026. The event will bring together leading experts in stochastic analysis and heat kernel methods to share recent advances, discuss connections between analysis and probability, and develop new collaborations on diffusion and stochastic dynamics on complex spaces. The conference will promote international collaboration and offer strong visibility for graduate students and postdoctoral researchers, helping connect researchers across career stages and areas of expertise. Heat semigroups provide a natural framework connecting probability, geometry, and functional analysis. Recent work has linked heat kernel methods to curvature-dimension conditions, sub-Riemannian geometry, spectral theory, and analysis on singular or rough spaces. At the same time, stochastic techniques such as SPDEs and rough path theory have advanced the study of nonlinear phenomena and scaling limits in complex systems. The conference will highlight both foundational developments and new applications, with particular emphasis on the interaction between probabilistic and analytic approaches. The website of the conference is: https://conferences.au.dk/saa-heat-kernels-2026. 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 · 2026-06

Abstract: Colorectal cancer ranks second in cancer-related mortality for both men and women in the United States. Nearly half of colorectal cancer patients exhibit aberrant activation of the MAPK/ERK pathway due to gain-of-function mutations in the BRAF and KRAS genes. Among these patients, those with the BRAF V600E mutation have the poorest prognosis. Therapeutic combinations of FDA-approved BRAF inhibitors with other agents targeting the MAPK/ERK pathway marginally increase survival, suggesting that blocking this pathway activates prosurvival mechanisms in colorectal cancer cells. The goal of this proposal is to identify these prosurvival mechanisms and design rational therapeutic combinations to enhance the efficacy of BRAF inhibitors by blocking these mechanisms in colorectal cancer cells harboring the BRAF V600E mutation. Our preliminary studies identified that inhibition of the MAPK/ERK pathway promotes mitochondrial fusion. We recently discovered that increased mitochondrial fusion acts as a prosurvival mechanism in Acute Myeloid Leukemia cells by enhancing mitophagy, an autophagic process that clears damaged mitochondria in response to BH3 mimetics treatment. Promisingly, pharmacologic inhibition of mitochondrial fusion combined with BRAF inhibitors synergistically eradicated colorectal cancer cells with BRAF V600E mutations. The proposed work will dissect the contribution of mitochondrial fusion to the survival of colorectal cancer cells upon treatment with MAPK/ERK pathway inhibitors and assess the therapeutic potential of targeting mitochondrial fusion in murine and organoid models of colorectal cancer. Additionally, this proposal outlines my career development plan for transitioning into a successful independent investigator. This plan includes participation in scientific and career development meetings, workshops, and coursework to gain more experience in cancer and chemical biology, and to hone my mentorship and leadership skills. Collectively, this research award is expected to profoundly impact the treatment of colorectal cancer, providing conceptual groundwork, preliminary data, and experimental tools that will lay the foundation for developing clinical trials combining mitochondrial fusion inhibitors with existing BRAF inhibitors and for launching my career as an independent investigator.

NIH Research Projects · FY 2026 · 2026-05

PROJECT SUMMARY/ABSTRACT We are requesting funds to upgrade our current FUJIFILM VisualSonics Vevo 3100 ultrasound system to the advanced Vevo F2 imaging system. The Vevo F2 will be housed within the Weldon School of Biomedical Engineering at Purdue University and will support over 1,900 hours of accessible user time per year for 11 Major Users and 7 Other Users. This upgrade is essential for our NIH-funded research community, which spans cardiovascular, gastrointestinal, renal, musculoskeletal, and oncological research. The Vevo F2 system offers significantly improved imaging performance over the Vevo 3100, including higher acquisition speeds, an expanded frequency range (71–1 MHz), and new imaging modes such as 4D Color Doppler, Vevo Strain 2.0, and elastography, etc. These capabilities will support imaging needs across animal models ranging from embryonic zebrafish to large animal systems, enabling precise, real-time imaging of cardiovascular and organ function that the current Vevo 3100 system cannot achieve. The current Vevo 3100 (acquired in 2018) is now approaching its limits in both performance and capacity, with 80% average usage and a lack of support for advanced imaging modalities (i.e. elastography, lower frequencies, acoustic engineering software (VADA), and color Doppler 4D imaging). The proposed Vevo F2 will significantly improve image acquisition efficiency, reduce animal anesthesia time, and expand capabilities in strain quantification, large-animal imaging, and multi-user workflows. Our collaborative laboratories, both domestic and international, are already using the Vevo F2 system, enabling data sharing and methodological harmonization. A nine-member advisory committee will oversee fair access and scheduling for current and prospective users of the Vevo F2 system. Daily operations and user training will be led by a dedicated team of experienced scientists. The Weldon School of Biomedical Engineering at Purdue has committed multiyear support for service contracts and data storage. This shared Vevo F2 system will continue to support and enhance current and future NIH-funded research projects and catalyze new directions in biomedical research.

NSF Awards · FY 2026 · 2026-05

The 2026 Midwest Machine Learning Symposium (MMLS) will be held at Purdue University in West Lafayette, Indiana, on June 24–25, 2026. MMLS is an annual regional conference dedicated to fostering communication and collaboration among machine learning researchers at all career stages across the Midwest. The symposium brings together graduate students, faculty, and industry researchers for a program featuring plenary talks by senior researchers from across the country, invited talks primarily by junior and mid-career Midwestern faculty, poster presentations largely by graduate students, and panels and industry-focused events led by senior and industry researchers. MMLS aims to advance the frontiers of machine learning and artificial intelligence and their applications, laying the groundwork for future interdisciplinary and cross-institutional collaborations. The symposium elevates the profile of machine learning research across Midwestern institutions by strengthening a regional community centered on collaboration and mentorship. With particular emphasis on the development of junior researchers, including graduate students, postdoctoral scholars, and early-career faculty, MMLS fosters both educational and professional growth among its participants. 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 2026 · 2026-05

This REU Site award to Purdue University, located in West Lafayette, IN, will support the training of 10 students for 10 weeks during the summers of 2026-2028. Nurturing the careers of these beginning scientists with specialized training, mentored research experiences, and professional development will encourage pursuit of advanced degrees and research careers in academia, industry, and government, thereby reinforcing strong recruitment into STEM fields of study including structural and computational biology and biophysics. Studies show that students who actively engage in undergraduate research within life sciences perform at higher levels within the classroom and are more likely to pursue advanced degrees. This training program will provide advanced training for students who will be part of a more highly trained and diverse workforce within a STEM discipline that will enhance U.S. global competitiveness. Students will learn how research is conducted, the ethical and responsible conduct of research, how to disseminate their results to a general audience, and will present the results of their work at scientific conferences. Assessment of the program will be performed using SALG and URSSA online tools. Students should apply to the REU site using NSF ETAP (Education and Training Application: https://etap.nsf.gov). The training students will receive is aligned with the NSF priorities in Biotechnology and AI. This REU program is designed to establish a framework that integrates research projects aimed at studying the structure and function of macromolecules with in-depth workshops designed to provide theoretical and practical training in contemporary structural biology techniques and biophysical methods. Students will gain training in methods such as computational biology, X-ray crystallography, nuclear magnetic resonance spectroscopy, and cryo-electron microscopy. Each trainee will leave the program with a solid foundation and understanding of the various structural biology methods regardless of the expertise within their host lab. These discovery-driven studies provide the molecular blueprint for the atomic structures of biomolecules which fuels subsequent hypothesis-based research. Projects included in this proposal focus on the structure and function of membrane protein machineries involved in signaling and nutrient import, viruses and viral enzymes, cancer-related proteins, CRISPR machinery, and transporter proteins. This REU program is hosted by the Department of Biological Sciences. Students that are in their sophomore or junior year are encouraged to apply. 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 2026 · 2026-05

A passive three-dimensional (3D) imager estimates the distance of objects from photographs captured without emitting light into the environment. Compared with active 3D technologies such as Light Detection and Ranging (LiDAR), passive 3D imaging offers important advantages in covertness, energy efficiency, and hardware simplicity, making it particularly attractive for applications in national defense, scientific exploration, robotics, and wearable devices. Despite substantial progress in both hardware and software, existing passive 3D imagers remain inherently limited by a short operating range, high computational cost, poor performance in low-light conditions, and difficulty integrating additional imaging functions. This project will develop a new family of passive 3D imaging solutions that overcome these limitations. Collectively termed as Computational Passive 3D Imaging, these solutions perform specialized computations directly on naturally available environmental light using coordinated optics and algorithms. Preliminary results demonstrate clear advances in range, efficiency, low-light robustness, and integrability. Once fully developed, these technologies are expected to transform 3D perception for wearable systems, robots, drones, autonomous vehicles, underwater platforms, and space exploration systems. The project will also advance engineering education by creating and disseminating learning activities themed around cameras for students from middle school through the graduate level, strengthening their experimental skills essential to the next-generation workforce. This project will introduce a set of novel imaging modalities that substantially extend the limits of passive 3D imaging in range, power consumption, low-light robustness, and multifunctionality. These modalities integrate advanced optical elements, including metasurfaces, microelectromechanical systems, and programmable optics, with both physics-based and learning-based algorithms. They perform specialized modulations on the scene’s plenoptic function during capture and complementary computations after measurement. The project will investigate the mathematical models underlying these imaging modalities, the hardware elements, and the algorithms. It will also develop simulation frameworks and experimental prototypes to analyze and explore the empirical performance of these new imaging platforms. The theoretical and practical outcomes should enable the design of advanced computational imaging systems beyond the scope of this project. 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 2026 · 2026-05

This award is for the deployment of a mobile weather radar as part of the Purdue University “Students Of Purdue Observing Tornadic Thunderstorms for Research (SPOTTR)” summer course. Students apply what they have learned in the classroom to real-world conditions during a field experience in the U.S. Great Plains to observe severe weather. The mobile radar significantly enhances this training by providing hands-on experiences with advanced observational technology. The project helps build a skilled workforce that supports improved weather forecasting and hazard preparedness. The University of Oklahoma Rapid-scan X-Band Polarimetric mobile radar (RaXPol) will be deployed first at Purdue University for orientation, training, and calibration exercises to introduce students to this advanced weather radar. Data collected during the deployment will be used for research projects and culminate in student presentations and publications. 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 2026 · 2026-05

This Faculty Early Career Development Program (CAREER) project will support the development of contact-aware robots which would leverage touch in intelligent ways. Existing efforts in soft robotics have aimed to limit robot interaction with physical disturbances in the environment for operational efficiency and safety considerations. However, contact-rich tasks such as fruit picking in dense trees will require robots which can appropriately apply and withstand contact with as much adaptability and robustness as humans, all the while remaining safe to be nearby. This project will advance fundamental knowledge and tools of robots designed to work with, instead of against, contact by investigating how passive responses and reactions to touch can be designed in a soft robot arm to do useful tasks, such as avoiding damage, seeking support, and automatically grasping target objects. Soft continuum robots with contact awareness will provide robust and adaptable solutions for domains where labor shortages are growing or where the risk to human workers can be significant, such as healthcare, agriculture, hazardous inspection, and search-and-rescue. The educational activities will leverage soft robot design tools from the research project to develop hands-on kits that use cost effective components to help middle and high school students explore basic engineering and science topics. These kits are expected to deepen students’ knowledge of these topics as well as giving opportunities for graduate student trainees to engage in educational development. Ecological psychology frames the environment and the agent, such as a soft continuum robot, as a combined system. An agent's behavior emerges from the dynamics of the combined agent and environment system, where the interaction has sufficient information both for adaptive behavior to emerge and for a representation of the environment to be encoded. Within this framing, this project aims to create passive open-loop behaviors, reactive local-loop behaviors, and integrated closed-loop behaviors which can understand and leverage the encoded information of compliant contacts with an environment. To achieve these goals, this project will focus on how pneumatic logic and geometric design variation, supported by physics-informed neural surrogate models, can be used to create design spaces with varying active and reactive contact responses. These responses will then be leveraged by contact-aware planning strategies with combined environment/robot transition dynamics to allow computational intelligence to leverage the designed mechanical intelligence. 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 2026 · 2026-05

This Faculty Early Career Development Program (CAREER) award supports research to determine how matrix-bound water contributes to the bone’s ability to carry load and resist fracture. Although bone fractures are common and costly, most tools used to estimate fracture risk rely primarily on measurements of mineral content and overlook the role of water and collagen in the bone matrix. This project will examine how matrix-bound water changes across the bone structure during aging and reduced mechanical use; and identify when and where water loss begins to impact mechanical performance. By combining advanced magnetic resonance imaging, experimental mechanics, and artificial intelligence (AI)–based analysis of spatial patterns, the research will lay the groundwork for imaging and computational models that more accurately reflect how bone fails. In parallel, a remote, scaffolded research program will provide undergraduate students, including students balancing work, caregiving, or remote study, with hands-on experience in biomedical imaging and modeling. Together, these efforts advance the fundamental understanding of bone mechanobiology, while equipping students with practical skills in imaging, data analysis, and computational modeling and advancing the modern engineering workforce. This project will determine how spatially localized changes in matrix-bound water compromise bone mechanical performance using spectroscopy, water-sensitive imaging, and mechanical testing. These measurements will be combined to develop and validate multiscale, imaging-based finite element models that incorporate dynamic, water-sensitive matrix properties. Two coordinated research efforts will be conducted. The first will quantify when and where matrix-bound water loss leads to declines in toughness and viscoelasticity, independent of mineral or collagen degradation, through complementary studies in animal and human bone that establish spatially resolved composition–mechanics relationships. The second will apply artificial intelligence methods to extract texture and heterogeneity features from water-sensitive imaging data and integrate these features into finite element models to improve prediction of local failure risk. Model predictions will be benchmarked against experimental mechanical outcomes and disseminated through an open imaging and modeling atlas. This work advances biomechanics and mechanobiology by defining when matrix hydration changes become mechanically meaningful and by enabling mechanics-based models of bone failure that reflect dynamic matrix composition rather than mineral structure and density alone. 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 · 2026-05

ABSTRACT To train the next generation of cancer researchers with interdisciplinary team science expertise, we propose establishing the Purdue Interdisciplinary Cancer Research Training (PICRT) program. PICRT leverages the research strength and breadth of the Purdue University Institute for Cancer Research (PICR), a longstanding NCI-designated basic science cancer center. The PICRT program fills a significant gap as, despite the strengths of PICR, there are currently no existing NIH training opportunities for Ph.D. students with a focus on cancer research at Purdue. PICRT is designed to train Ph.D. students in interdisciplinary knowledge, skills, and innovative critical thinking across cancer research domains. These competencies are essential for trainees to effectively collaborate in multidisciplinary teams, driving progress from basic discovery and prevention to treatment and clinical application. The objectives for the proposed training program are: 1) enhance knowledge and skills in interdisciplinary cancer research; 2) deliver enhanced training to increase interdisciplinary and collaborative cancer team science competence; 3) increase trainee cancer research career awareness and provide professional development opportunities and enhanced effective leadership skills. The PICRT program includes 35 preceptors distributed across the three broad interdisciplinary research Scientific Programs of PICR: Cell Identity and Signaling (CIS), Drug Delivery and Molecular Sensing (DDMS), and Targets, Structures, and Drugs (TSD). The composition of the preceptor pool spans 12 departments, including Biology, Industrial and Molecular Pharmaceutics, Biochemistry, and Chemical Engineering. The proposed program is led by a multi-PI team of leaders in cancer research with diverse research backgrounds and approaches, long-standing strengths in graduate education and administration, and substantial institutional support. Our program novelty comes from overcoming barriers that insulate researchers from working together as interdisciplinary research teams, using previous successful curricular development to enhance skills in working in interdisciplinary teams. In addition to more traditional training program components, novel aspects of PICRT are comprised of a course that includes training in interdisciplinary team science, clinical cancer applications, an embedding program in different interdisciplinary laboratories, opportunities to organize PICR events, and leadership experience in cancer-focused service learning/community projects. Our vision is that these broadly trained basic cancer researchers will be well suited to work in teams to identify, develop, and implement cancer prevention and treatment strategies in partnerships with clinicians. The robust interdisciplinary research environment at PICR, the excellent preceptor and Ph.D. pool, the strong leadership team, and the substantial institutional support poise us well to successfully train the next generation of highly competent cancer researchers and fill the gap at Purdue and the overall cancer research workforce.