University Of California Berkeley

NSF Awards · FY 2025 · 2025-09

This I-Corps project focuses on the development of a wearable biosensor that continuously measures molecular biomarkers in the human body. Current approaches to biochemical monitoring rely on infrequent sampling and lab-based analysis, which miss rapid changes and provide limited insight into dynamic physiological processes. The technology introduces a new sensing platform that captures real-time biochemical data noninvasively and with high sensitivity. By enabling continuous monitoring, the technology supports advances in health tracking, early disease detection, and personalized care. This approach promotes scientific progress and public well-being by making biochemical information more accessible, timely, and actionable. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on the development of electrochemical biosensors that use structure-switching deoxyribonucleic acid integrated with custom low-power, low-noise complementary metal-oxide-semiconductor circuits. The sensors convert molecular binding events into electrical signals, allowing real-time quantification of multiple biomarkers at relevant concentrations. The platform includes wearable and point-of-care formats designed for small sample volumes and wireless operation. This innovation advances biosensing by combining biochemical specificity with scalable electronics, enabling new applications in longitudinal health monitoring and mobile diagnostics. 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 I-Corps project focuses on proteins containing functional materials that are compatible with existing manufacturing infrastructure, accelerating their commercial adoption. Embedding proteins into plastics creates a new class of hybrid materials, bioactive plastics. Bioactive plastics combine the unique functions of biology with the production throughput and versatile forms of synthetic materials. For instance, dispersing nanoscale enzymes within biodegradable polymers can lead to fully compostable plastics. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This solution is based on factors that govern enzyme stabilities outside of their native environment, e.g. inside of plastic matrices and during the polymer processing. Specifically, the stabilization technology is guided by protein conformation, the activity of the short-range interactions, and the chemical and physical characteristics of the surrounding media. Here, embedded enzymes convert polymers into small molecules under industrial composting conditions. Protein integrity is maintained during plastic manufacturing processes like 3D printing, pelletizing, film casting, and molding. This enables the downstream processing of bioactive plastics for health and sustainability applications, such as surgical implants with embedded protein therapeutics and plastics programmed with custom degradation profiles for specific real-world environments (e.g., composts/landfills, oceans, or re/upcycling facilities). 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

How can we interpret results from complex machine learning algorithms? How can we mitigate the risks associated with using such models for policy decisions? This project addresses fundamental challenges in deriving valid, reliable, and interpretable causal conclusions from complex data using modern machine learning tools. As machine learning becomes increasingly integral to disciplines such as medicine, economics, education, and the social sciences, the demand for causal insight --- beyond predictive accuracy --- has become more pressing. Yet many machine learning algorithms function as “black boxes”, offering limited transparency and lacking rigorous frameworks for replicability and uncertainty quantification. This project aims to establish a theoretical foundation for causal learning that makes outputs from machine learning explainable, statistically sound, and actionable in real-world decision-making. The work is complemented by educational and outreach activities that promote understanding of causal reasoning among students and the broader public. Planned efforts include public lectures, collaborations with K–12 educators, and integration of research findings into university curricula. Collaborative partnerships with institutions such as Microsoft, Eli Lilly, and the Fred Hutchinson Cancer Center will help translate methodological advances into impactful scientific and societal applications. Technically, the project advances causal learning through three interrelated aims. (1) It develops methods for imputing unobserved counterfactual outcomes --- the hypothetical “what if” scenarios that form the core of causal reasoning --- by integrating flexible machine learning models with statistical principles to preserve both interpretability and rigor. (2) It promotes design-based approaches for quantifying uncertainty, particularly in settings where treatments are assigned randomly or pseudo-randomly via permutations. These methods isolate uncertainty from treatment allocation mechanisms, complementing model-based inference. (3) The project builds a statistical framework for finite-population inference, extending traditional inference techniques beyond super-population assumptions. By drawing on tools from empirical process theory and random matrix theory, the framework provides robust inferential guarantees in realistic data settings where independence and large-sample assumptions fail. Together, these contributions will advance the theory and practice of causal learning, bridging machine learning and statistics to improve both scientific understanding and data-informed decision-making. 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

Glaucoma is chronic degeneration of the retina and optic nerves. However, the mechanisms by which ganglion cells in the retina communicate with cells in the myelinated optic nerve during ocular hypertension (OHT) are unclear. We recently discovered a resident neuroprotective retinal astrocyte- ganglion cell lipoxin circuit that is impaired during retinal stress, including OHT. Homeostatic astrocytes produce two endogenous lipoxins, Lipoxin A4 and Lipoxin B4, which act directly on retinal ganglion cells (RGCs). LXB4 is the most potent lipoxin in the retina, and it increases the survival and function of RGCs in OHT-induced neuropathy. To elucidate mechanism of actions of LXB4, we employed single-cell RNA sequencing, morphological characterization, and RNA sequencing to identify microglia as a novel cellular target for LXB4. We discovered that OHT induces distinct and temporally defined microglial functional phenotypes during the time course of sustained OHT in both the retina and myelinated optic nerve. Unexpectedly, microglial expression of CD74, a marker of disease-associated microglia (DAM), was only induced in the optic nerve but not in retinal microglia. LXB4 treatment during OHT shifted the optic nerve microglia toward homeostatic morphology and downregulated the expression of CD74. These findings identified microglia as a new LXB4 target in the optic nerve. LXB4 maintenance of optic nerve microglia and inhibition of the DAM phenotype are potential mechanisms for LXB4 neuroprotection. Hence, we hypothesized that the induction of CD74+ DAM in the optic nerve during OHT is a key signature of degeneration, and its regulation by LXB4 is neuroprotective mechanism against OHT-induced neuropathy. Aim-I during K99 phase will define how CD74+ DAM is regulated by LXB4 in the optic nerve during chronic OHT and identify targets and mechanisms specific to CD74+ DAMs in the optic nerve. For the R00 phase, in Aim-II, I intend to study and investigate how microglia in the remote distal myelinated optic nerve sense the insult to RGCs. In Aim III, I intend to study and investigate the role of optic nerve-specific DAMs driving the degeneration of axons. I will use spatial transcriptomics of optic nerve to define the cell -specific targets and differential proteomics approach to study changes in pathways and mediators in the optic nerves after OHT insult. To address the role of CD74+ DAM in the optic nerve during OHT, CD74 knockout mice will be used to study phenotypic, transcriptomic, and signal transduction regulation. In addition, to understand the crosstalk of glial cells in the optic nerve during OHT insult, I will use spatial transcriptomics based bioinformatic methods and electron microscopy to determine its effect on myelin health. Understanding the role of CD74+ DAM in the myelinated optic nerve during chronic OHT will deepen our knowledge of neurodegeneration mechanisms and open new paths for therapeutic interventions.

NIH Research Projects · FY 2025 · 2025-09

Project Summary Over the past two decades, point-of-care (POC) systems enabled detection of biomarkers at or near the patient, bypassing the need to transport samples to a centralized lab, thereby enabling rapid molecular diagnoses. A key innovation that enabled this technology is the integration of sample preparation processes into a miniaturized, disposable “lab-on-a-chip” device. However, most current POC systems still require an external instrument for automated processing and detection, increasing the system size, complexity, and cost. This makes such systems impractical for low-resource settings. To address this need, we propose a miniaturized, disposable CMOS-based POC system that can be ”dropped” into the sample, measure the concentrations of multiple analytes, and wirelessly transmit the results to a smartphone. Our system is powered by two key innovations: aptamer switches that generate binding-specific electrochemical-signals in complex samples, without the need for any sample preparation, and a millimeter-sized single-chip CMOS electronics that houses and measures multiple of these aptamer switches and wirelessly transmits the results to an external smartphone reader. We will specifically tailor our diagnostic system to measure two urine metabolites crucial for the early detection of preeclampsia, a prominent contributor to maternal mortality in developing nations. The project objective will be achieved by four specific aims: (1) Aim I will implement the CMOS chip, featuring signal-enhanced square-wave voltcoulometry-based electrochemical readout electronics, wireless power- harvesting circuits, and data transfer protocols, all incorporated in mature low-cost 180-nm CMOS technology. (2) Aim II involves selecting and screening novel aptamers targeting sialyllactose and uric acid ribonucleotides, exploiting our established Particle Display and Aptamer Array technologies. Concurrently, high-throughput switch domain screening will engineer aptamers into structure-changing switches, allowing direct in-sample detection without reagent requirements. (3) Aim III emphasizes system integration, entailing post-CMOS fabrication of nanoporous electrodes for additional performance enhancements and developing an integrated wireless powering reader. (4) Aim IV will optimize this new point-of-care workflow and validate detection in urine samples. Moreover, leveraging the multiplexed detection capability, we will develop deep-learning techniques to enhance concentration mapping accuracy, particularly in fluctuating environmental conditions. This project holds significant promise; its successful completion could transform future diagnostics, particularly benefiting resource- constrained environments. These settings often face high maternal morbidity and mortality rates due to prevalent health complications and medical and laboratory infrastructure insufficiency.

NIH Research Projects · FY 2025 · 2025-09

Project Summary / Abstract Solubility is a crucial factor in pharmaceutical drug discovery and development, directly impacting bioavailability, efficacy, and formulation. However, experimental solubility assays are often complex and resource-intensive, while informatics-based approaches are hindered by the lack of comprehensive, high-quality datasets. Atomistic simulations, though promising, are limited by inadequate interatomic potentials, challenges in free energy estimation, and the polymorphism of molecular crystals. Our lab focuses on designing and applying atomistic simulations, statistical mechanics, and machine learning methods to predict the properties of materials and molecules. Our long-term vision is to create a computational platform that transforms drug discovery by enabling reliable predictions of key molecular properties. Over the next five years, we will develop a robust and automatable computational framework to predict the solubility of drug-related molecular crystals. This effort will involve creating efficient machine learning potentials (MLPs) based on quantum mechanical calculations, designed for broad applicability to organic systems of biological relevance. We will leverage these MLPs, along with generative models, to enhance crystal structure predictions. We will also build an automated workflow for high-throughput solubility computations across a diverse set of molecular systems under consistent conditions. Additionally, we will develop data-driven models that improve the accuracy of informatics-based solubility predictions. The computational framework will be validated against publicly available solubility datasets and further tested using blinded solubility data for newly synthesized compounds. This will demonstrate the accuracy and utility of the framework in real-world drug discovery contexts.

NIH Research Projects · FY 2025 · 2025-09

ABSTRACT Shigella infection causes shigellosis, a gastroenteric illness characterized by bloody diarrhea. Shigella infection affects over 80-165 million individuals per year worldwide, resulting in approximately 600,000 deaths. Countries with developing sanitation systems are heavily affected, with young children being primary victims of the disease. Currently, there are no vaccines or prophylactic options for Shigellosis. Although drugs like ciprofloxacin are often used to treat Shigella infection, antibiotic-resistant strains are on the rise, resulting in increased prevalence and mortality. Susceptibility to Shigella infection is influenced by both the immune system and the gut microbiome, the complex community of resident microbes in the intestine. However, the mechanism of protection is not well understood. Recently, data from our lab and others show the important role of the gut microbiome in protection against Shigella infection. For example, Bacteroides thetaiotaomicron (also known as B. theta) of phylum Bacteroidetes, a prominent member of the human gut microbiome, produces outer membrane vesicles that reduce Shigella virulence in culture. We have also identified that excreted products from B. thetaiotaomicron and the gut bacterium Enterococcus faecalis can decrease growth of Shigella. These data reveal the gut microbiome’s propensity to act as therapeutic agents. To continue investigating the role of the gut microbiome on Shigellosis based on these promising findings, I propose that by using bacterial culturing and innovative mouse models, we will determine how “healthy” gut bacteria can mitigate Shigella infection. In this proposed study, I will: 1) examine the effects of various gut bacteria on growth and virulence of Shigella through growth curve assays and a fluorescent reporter plasmid for virulence; 2) utilize the gnotobiotic mouse model to determine effects of defined gut bacterial communities on Shigella colonization, cell invasion, and disease outcomes. Narrowing which bacteria help mitigate infection can contribute to the discovery of protective microbes and inspire the development of probiotic candidates for disease prophylaxis for high-risk individuals. Training: I am well suited to conduct these studies due to my expertise in studying prophylactic options for bacteria-caused gastrointestinal diseases such as C. difficile infection. Recently, I have translated my previous skills into studying probiotic solutions for Shigellosis. In my preliminary research, I have determined a handful of Bacteroidetes-related bacteria that can reduce Shigella growth. I have also recently been able to establish Shigella infection in the germ-free mouse. This proposal will build on my prior knowledge of infectious gastrointestinal diseases and my preliminary research findings on the gut microbiome effects on Shigella. Moreover, this research will lay the groundwork for my overarching long-term goal of establishing my own interdisciplinary research group at a top research university focusing on defining the role of the gut microbiome in various debilitating human disorders and diseases. This research training plan will prepare me to develop into an independent investigator to lead impactful research that integrates microbiology, molecular biology, and biochemistry. With the guidance of my sponsor and mentorship committee, I will be well-equipped to meet these goals. Dr. Ashley Wolf’s lab has expertise in infectious diseases, gut microbiome, gnotobiotic animal models that, in conjunction with resources and collaborations at UC Berkeley, will provide an environment for success for me to conduct this research.

NIH Research Projects · FY 2025 · 2025-09

Abstract (250 Words) Multimodal artificial intelligence (MAI) systems hold significant promise for improving diagnostic accuracy in diverse clinical domains, particularly in dermatology, where various data sources such as clinical images, dermoscopy, pathology, and text are integral to diagnosis. However, stakeholders, including clinicians, have minimal involvement in the development of medical AI, which are often entirely designed and deployed by engineers as “black-box” systems. This opaque and one-way approach to AI deployment hinders accountability and lacks a feedback mechanism for stakeholders to correct AI errors. In dermatology, where single-modality AI often struggles with spurious correlations and lack of training data for underrepresented patient groups, the adoption of multimodal models raises concerns about exacerbating these errors and biases. To address this, we propose "Participatory MAI" for skin disease—a framework for co-design in which dermatologists directly intervene in MAI models to correct errors throughout their deployment. Our approach involves developing MAIs with novel editability capabilities that enable stakeholders to apply explicit modification to the MAI behavior using interpretable natural language instructions. The project will contribute new methods, models and datasets for Participatory MAI in dermatology. It will deliver (1) the first algorithms for multimodal model editing, (2) a proof-of-concept editable MAI for skin disease (DermaCLIP), and (3) a first publicly-accessible multimodal dataset for dermatology (Multi-Skin) alongside a pilot study evaluating our Participatory MAI approach. Integrating editing capabilities into MAIs marks a shift from opaque data-driven fine-tuning to transparent participatory fine-tuning under human oversight. Outcomes of this project are widely applicable across various clinical domains and modalities.

NSF Awards · FY 2025 · 2025-09

Humans use memories of the past to figure out where to pay attention and what is most relevant to reach their current goals. For example, when driving to work, an individual can use their memory for specific streets to direct their attention to intersections that they know are tricky or dangerous. This process of “memory-guided attention” helps individuals behave efficiently and accurately even if they are in a rush. Memories, however, can be a double-edged sword. Because many street intersections can look similar to one another, confusing one intersection for another can lead to costly mistakes. This project aims to determine how the brain enables humans to minimize confusion between similar memories to direct their attention accurately and precisely. Understanding how memories are used for rapid decisions about where and how to pay attention has implications for education and artificial intelligence models. For educational settings, the project has potential to inform how students can be taught to avoid confusing similar memories of related problems and concepts, which could help students pay attention in the right way at the right time. More broadly, understanding how the brain solves this problem could inspire new ways that artificial intelligence models can be changed to avoid confusing memories for similar events, and to predict human attention more accurately. In addition to the research, this project includes research training and mentoring for high school, undergraduate, and graduate students in cognitive neuroscience. This project’s main focus is to determine how the brain’s memory systems help prevent competition, and promote cooperation, between memories to guide attention. This project focuses on a key brain region that is critical for building new memories and retrieving old ones: the hippocampus. The main prediction is that the hippocampus helps minimize confusion between similar memories by forming differentiated neural representations of them over time, and that this process helps prevent errors in memory-guided attention. This prediction is tested in using multiple methods: functional magnetic resonance imaging, eye tracking, measures of behavioral accuracy and response times, and sophisticated statistical models of brain activity and behavior. The project focuses on memories at different timescales. Aim 1 focuses on memories that were acquired some time ago and retained in long-term memory. Aim 2 examines interactions between these long-term memories and working memories acquired several seconds in the past. In Aim 1, the goal is to determine how the hippocampus prevents similar long-term memories from competing to guide attention. It tests the hypothesis that competition is initially overcome by brain systems involved in effortful control of behavior, with the hippocampus taking over to guide attention with additional experience. In Aim 2, the goal is to determine how working memories and long-term memories may be represented distinctly in the brain and over time in behavior to prevent them from competing with each other and helping them cooperate instead. Together, the overall goal is to uncover the powerful ways that the mind and brain minimize competition and promote cooperation between memories to influence attention. 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

The essence of a many-body random system is often captured by its scaling limit, the structure arising when the system’s characteristic scales of space and time are brought to unit order by a linear transformation. And it is often profitable to introduce a dynamical enhancement of the system, whose snapshot at any given moment of time is the system itself, because the enhancement may have attractive properties that the original lacks, or because the behavior of the enhanced system at exceptional moments reflects aspects of the system’s scaling limit. The proposal addresses several random scaling limits and random dynamic enhancements of statistical mechanical systems. This project involves graduate student training. Principal directions of inquiry include study of the random scaled dynamics of a random walker whose attempted linear progress is frustrated by hard obstacles that clutter the route forward; the scale of time in a randomly evolving energy landscape that witnesses convergence to equilibrium of the depth of the deepest valley; and the scaled counter trajectory in random-turn games played on domains in Euclidean space. The proposal will develop and exploit basic tools from the theory of Ornstein-Zernike decay and subcritical connections, and discrete harmonic analysis. By dissemination, mentorship and collaboration, the PI will seek to ensure that the research enhances the mathematical experience and trajectory of junior researchers including graduate students via joint research to develop fundamental tools, and undergraduates via research study and coding projects. 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.

CIHR Grants and Awards · FY 202526 · 2025-09

Cannabis use in young adults is a significant and growing public health concern. North America has the highest prevalence of past-year cannabis use (14.5%), with young adults representing the largest demographic. Despite increased availability and social acceptance coinciding with legalization, frequent cannabis use in young adults poses serious public health concerns, such as mental health challenges, cognitive impairments, and difficulties in daily functioning. However, current treatments for cannabis use are limited in both accessibility and effectiveness, creating an urgent need for innovative and effective solutions. My research aims to address this critical gap by evaluating a novel, brief, digital intervention to help young adults reduce cannabis cravings and use. Regulation of Craving Training (ROC-T) is a mechanism-focused intervention based on Cognitive Behavioral Therapy (CBT) and is effective in reducing use of alcohol and other drugs, but it has not yet been evaluated for cannabis use. This study will recruit 60 young adults who use cannabis weekly and are motivated to reduce or quit their use. Participants will be randomly assigned to either the ROC-T or a control intervention, with all sessions conducted online, making this a highly accessible and potentially scalable intervention. The proposed study will examine whether ROC-T reduces cannabis use and cravings more effectively than a control intervention, and whether changes in craving directly relate to reductions in cannabis use. If successful, this research will lead to a pioneering, evidence-based, digital intervention that reaches young adults who might not otherwise have access to effective treatment, including those from traditionally underserved groups. This work represents a significant step towards improving mental health equity for young adults by providing a scalable approach to reduce cannabis-related harm, ultimately benefiting both individual well-being and public health outcomes. Keywords: CANNABIS USE; CRAVING; REGULATION OF CRAVING; COGNITIVE REAPPRAISAL; YOUNG ADULTS

NSF Awards · FY 2025 · 2025-09

In this project, artificial intelligence (AI) will be used to design new sustainable polymeric materials with a range of properties and that can be recycled without the need for costly and inefficient separation from mixed waste streams. Today’s plastic waste challenge exists at a scale of megatons per day across tens of thousands of applications and products. The researchers will create new types of depolymerizable plastics derived from simple feedstocks, and they will develop physics-informed AI models to aid design of these plastics such that they meet a variety of product specifications across a wide range of properties. Ultimately, this approach enables products of all different types, functions, and lifetimes to be integrated into a single recycling stream and accelerates their discovery-to-use timeline. The results and methods developed by this research will be publicly accessible for industrial benchmarking and include code and tutorials for users to perform AI-guided design on their own materials. Through this research, a new generation of scientists will be trained to work at the emerging intersection of polymer materials design and AI model development and use. With this award, the project will develop physics-informed AI and synthesize architecturally varied and deconstructable (ADD) polymers by cationic ring-opening polymerization (CROP) with controlled chain length, branching, and dynamic bond incorporation. This work will create new synthetic strategies to control chain end and side chain functionality, branch type and frequency, and dynamic bond incorporation for polymers produced by CROP. Using polyacetals and polyethers synthesized from a select few monomers, these complex molecular architectures will be linked to key properties using physics-informed AI, which both describes polymer architecture through sets of probability distributions and incorporates theoretical estimates of structure-property relationships. This physics-informed AI will be iteratively improved through active learning approaches and subsequently used to perform inverse design for the creation of new ADD polymers with targeted properties within specified tolerances that will be experimentally validated against industry benchmarks. This Molecular Foundations for Sustainability: Sustainable Polymers Enabled by Emerging Data Analytics (MFS-SPEED) award is co-funded by the NSF through the Division of Chemistry (CHE), the Directorate for Mathematical and Physical Sciences (MPS), and the Division of Innovation and Technology Ecosystems (ITE) in the Directorate for Technology, Innovation, and Partnerships (TIP). Additional MFS-SPEED funding is provided by Procter & Gamble, PepsiCo, Dow, BASF, and IBM. 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

Soil is teeming with bacteria and other microbes that drive the flow of nutrients on Earth, absorbing and releasing greenhouse gases like carbon dioxide (CO2). This nutrient flow depends on the ways different microbial species interact with each other and with other organisms including plants. As plant roots grow through soil, they release chemicals that feed the surrounding microbes in the region around roots known as the rhizosphere. In the rhizosphere, microbes also interact with each other by sharing micronutrients such as vitamins, but how micronutrient sharing impacts microbial community dynamics remains unknown. To study the links between microbial growth, micronutrient sharing, and carbon release, researchers will assemble a model community of bacteria within customized chambers that mimic the chemical environment of the rhizosphere. This novel experimental system will serve as a model for other scientists studying similar processes. Results of this study will be used to test existing computational models that predict how soil microbial communities influence carbon cycling. In addition to training students and postdoctoral scholars, this project will involve local 6th and 7th graders and their teachers to bring authentic research experiences to classrooms and to train the next generation of scientists. Microbial community assembly in the rhizosphere is shaped by microbial interactions, functional traits, and changing root chemistry. These factors modulate carbon use efficiency (CUE) – the proportion of carbon taken up by microbes and assimilated as biomass rather than released by respiration as CO2. Microbial CUE is key to forming stable soil organic matter. While it is known that macronutrient cycling (e.g. carbon, nitrogen, or phosphorus) contributes to microbial community assembly, the roles of micronutrient-sharing interactions are not well understood. This project will use corrinoids – vitamin B12 and related micronutrients – as a model to characterize the functional roles of micronutrient-sharing bacteria as well as their links to macronutrient cycling and important biogeochemical processes in the rhizosphere. Research activities include (1) prediction of how corrinoids influence the ecology of rhizosphere bacteria by assessing corrinoid-dependent interactions with bacteria grown individually and in combination with others; (2) development of a model rhizosphere bacterial community to study how corrinoids alter their ecology, the availability of macronutrients, and together how these interactions shape bacterial succession; and (3) construction of gene knockouts as well as community dropouts to examine how disruptions in micronutrient interactions impact bacterial community dynamics. Researchers will modify an existing dynamic energy budget model to incorporate micronutrient interactions in order to show how these interactions alter microbial community phenotypes, such as CUE. 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 · 2025-09

Abstract/Project Summary Cigarette smoking is the leading preventable cause of disease, disability, and death1. Unfortunately, current smoking cessation treatments are suboptimal, and innovative interventions are needed, particularly those that are accessible and scalable (e.g., brief, digital) and target core mechanisms. Craving2-5 and the ability to regulate craving6-8 are two core mechanisms implicated in smoking. Although pharmacotherapies (e.g., nicotine replacement therapy; NRT) reduce craving and smoking9, 10, they have modest effects on quit rates11, 12 – suggesting pharmacotherapy alone is inadequate. Critically, combining behavioral and pharmacotherapy approaches for smoking cessation is more effective than either alone9. Combining anti-craving pharmacotherapy with a targeted, mechanism-focused behavioral intervention for enhancing regulation of craving may be highly promising, yet is understudied. We developed Regulation of Craving Training (ROC-T)13-17, a brief, digital intervention providing targeted, intensive training in skills to regulate craving, via repeated practice of a skill while viewing smoking-related images that elicit craving. There are two versions of ROC-T: (1) Reappraisal ROC-T (RROC-T)13, 14: training in reappraisal – reframing smoking as an undesired behavior/non-smoking as desired (based on cognitive treatments18, 19); and (2) Mindfulness ROC-T (MROC-T)17, 20: training in mindful- acceptance – accepting craving as it is and letting it pass (based on mindfulness treatments21, 22). Our NIDA- funded Stage 1B trial (n=92) showed that both versions vs. an assessment-only control led to large-sized reductions in craving and smoking23. Given its brief, digital format, and promising results thus far, ROC-T would be ideal as a preparation intervention24-28 added to the beginning of standard care (SC) – NRT and counseling – and prior to quitting. Indeed, individuals with lower craving immediately prior to29-37 and after31, 35, 38-43 the quit date are more likely to stay abstinent from smoking in the long run. We propose a Stage 2 efficacy trial44, 45 with 882 adult cigarette smokers who will receive SC and will be randomized to (1) RROC-T, (2) MROC-T, or (3) Sham ROC-T (control) as adjunctive preparation interventions that are integrated into SC. Each ROC-T includes 6x30-min modules completed during weeks 1-4 in preparation for a quit date at the start of week 5. Sham ROC-T includes neutral images with no skills training. SC includes 7x15-min virtual counseling sessions9, 46 and NRT (patch plus lozenge or gum47, 48; initiated on the quit date). The primary outcome is CO-verified 7-day point prevalence abstinence (PPA) 6-months post-quit. Abstinence will be confirmed with video-based CO testing. To assess real-time craving and skill use in daily life, participants will complete daily ecological momentary assessment (EMA) for 1-week at baseline, and 1- week pre-quit through 4-weeks post-quit. This project has the potential to significantly advance mechanism-targeted, accessible interventions for smoking cessation.

NSF Awards · FY 2025 · 2025-09

This project aims to develop a new artificial intelligence system that works alongside mathematicians to tackle problems that have resisted solutions for nearly a century. Recent advances in large language models can generate creative insights and partial reasoning steps, but they often make mistakes and cannot guarantee correctness. In contrast, traditional tools for verifying mathematical proofs offer rigorous guarantees but are not well-suited for automatically navigating the vast search spaces involved in complex mathematical discovery. This research combines the strengths of both approaches: using AI to explore promising ideas and using formal logic to rigorously verify and refine them. As a high-impact test case, the team will focus on the Hadamard Conjecture, a longstanding open problem with applications in quantum error correction, communication systems, and coding theory. The project will also produce open-source tools, educational materials, and outreach programs to broaden participation in advanced mathematics and AI. The research introduces a unified framework with three key components: (1) a self-evolving reasoning pipeline that uses synthetic data to guide exploration of promising matrix constructions; (2) chain-of-thought and curriculum learning to help AI decompose complex mathematical tasks into simpler subproblems, integrate partial solutions, and generalize from simpler to more difficult problems; and (3) formal verification tools, such as Lean, integrated with preference alignment to ensure correctness and enable a self-improving system guided by symbolic proof signals. Together, these elements form a closed-loop system for scalable, trustworthy proof generation. Anticipated outcomes include new Hadamard matrix constructions, practical software for AI-assisted mathematics, and foundational advances in combining learning and logic for mathematical problem solving. 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

Unicode Consortium is a technical standards body comprised of industry partners that is responsible for developing standards for the universal character encoding for digital environments, including AI technologies. This project investigates how the Unicode standards are handling newly invented scripts called “neographies.” Whether a writing system is encoded in Unicode directly affects its ability to be incorporated into AI tools and other core text technologies. This research presents an opportunity to examine how technologists are making decisions that shape the digital environments and language policy. This project offers a systematic assessment of Unicode’s treatment of neographies. It involves analysis of proposals related to this class of writing systems, participant observation and interviews with Unicode standards-makers and neography proponents, and case studies of selected neographies from contrasting geographic and political contexts. This project aims to understand what kinds of evidence are considered in these cases, how eligibility criteria are applied across contexts, and how standards-makers navigate the line between technical and linguistic authority. The outputs of the project inform decision-making within technical standards bodies and contribute to shaping industry practices around multi-script and multilingual support. 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

Understanding the signaling between neurons, glial cells, and blood vessels during development is crucial for designing treatments for vasculature defects, such as the over-vascularization associated with retinopathy of prematurity. The proposal aims to determine the role of neural-glial-vascular signaling in vasculature development through two main objectives: Aim 1: Use quantitative approaches to characterize the vasculature development and to assess how this process is altered in mice with altered retinal waves. We also propose to use live imaging of vasculature growth to assess the nature of dividing vessel maturation. Aim 2: Establish the maturation of the glial-vascular unit using two approaches. First, we will use immunofluorescence to assess when the glial-vascular interface arises during development. Second, we will use two-photon calcium imaging to monitor the spontaneous calcium transients in glial processes associated with vasculature and establish their correlation with retinal waves. This research will enable the lab to test novel hypotheses regarding the role of signaling in vasculature development, which is critical for understanding how neural activity influences normal and pathological vasculature development.

NSF Awards · FY 2025 · 2025-09

With the support of the Macromolecular, Supramolecular and Nanochemistry Program in the Division of Chemistry, Professor Haimei Zheng of the University of California, Berkeley is studying copper-based nanocatalysts for carbon dioxide (CO2) conversion to useful fuels and chemicals. Copper-based bimetallic nanocatalysts have gained significant attention for their potential to improve the process of CO2 transformation; however, these catalysts do not remain unchanged during reactions—their surfaces often shift away from their original structures. Understanding how and why these changes occur is essential, as they can greatly affect how well the catalyst performs and how long it remains effective. This project aims to uncover how copper-based bimetallic catalysts transform during CO2 conversion using a powerful technique called in-situ electrochemical liquid cell transmission electron microscopy (EC-TEM). This is currently the only method capable of directly observing such atomic-scale changes in real time. By watching these transformations as they happen, the research team hopes to uncover how changes in structure and chemical bonding influence catalytic performance. The knowledge gained could help scientists design better, more efficient catalysts for energy applications. Beyond its scientific goals, the project will also provide valuable training opportunities for both graduate and undergraduate students. It will help prepare the next generation of scientists and engineers while contributing to the long-term competitiveness of U.S. technology and innovation. With the support of the Macromolecular, Supramolecular and Nanochemistry Program in the Division of Chemistry, Professor Haimei Zheng of the University of California, Berkeley is studying copper-based bimetallic nanocatalysts for electrochemical carbon dioxide reduction (CO2RR), focusing on model systems such as CuM (M = Pd, Ag, etc.) to elucidate atomic-scale restructuring processes using in-situ EC-TEM. The research will involve fabricating advanced electrochemical liquid cells with precisely controlled potentials for high-resolution, real-time imaging; performing EC-TEM studies on well-defined CuM catalysts with varied structural configurations to capture atomic and morphological transformations during CO2RR; and conducting complementary multimodal characterizations across multiple length scales, including low-dose cryo-EM imaging and spectroscopy to resolve the atomic structure and chemical bonding of reaction intermediates, as well as operando X-ray absorption spectroscopy (XAS) to monitor valence state evolution in catalyst ensembles. In parallel, the project will integrate theoretical modeling and electrochemical performance testing through established collaborations to elucidate restructuring mechanisms and correlate surface structure with CO2RR activity and selectivity. Collectively, these efforts are expected to yield critical mechanistic insights into the dynamic behavior of bimetallic electrocatalysts, ultimately guiding the rational design of next-generation CO2RR catalysts with enhanced efficiency, durability, and product selectivity. 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 this project, funded by the Chemical Mechanism, Function, and Properties Program of the Chemistry Division, Professors F. Dean Toste of the Department of Chemistry and Bryan D. McCloskey of the Department of Chemical and Biomolecular Engineering at University of California, Berkeley, alongside Professor Jacob S. Tracy of the Department of Chemistry at the University of West Florida, are developing strategies for creating, understanding, and optimizing new redox-active organic molecules to enable next-generation redox-flow and metal-air energy storage systems. The impacts of this work cover long duration energy storage systems, which enable overall electrical grid stability when incorporated alongside intermittent electricity generation, and ultrahigh capacity battery systems which aid heavy duty transportation applications. The project takes an interdisciplinary approach to the challenges inherent to these complex systems, combining synthetic organic and physical organic chemistry with electrochemistry and chemical engineering. Further, this project is the collaborative effort of a graduate-focused institution and an undergraduate-only chemistry department that addresses scientific training and education with impacts on students from K-12 all the way to the graduate level. Redox-active organic small molecules have an immense level of structural tunability to achieve various reduction potentials, stabilities to oxidation/reduction, and reactivities in solution and across interfaces. They have been used as active materials in nonaqueous organic redox flow batteries (N-ORFBs) and as soluble redox mediators (RMs) in metal-air (M-O2) batteries to enable or improve long-term, high-capacity performance in those systems, but holistic design rules and a mechanistic understanding of their decomposition pathways in either system remain limited. This project will take a physical organic chemistry approach to the design of both N-ORFB materials and M-O2 RMs. The broad thrusts of the RFB-focused work are (1) the design and characterization of highly cyclable, extreme-potential redox flow battery catholyte and anolyte materials and (2) development of a mechanistic understanding of how regioisomerism and intramolecular noncovalent interactions affect the voltages and long-term stability of RFB materials. For M-O2 batteries, the project focuses on (1) synthesizing new classes of mediators for Li-O2 batteries alongside mechanistic studies of their decomposition in model cell systems and (2) investigation of the fundamental electrochemistry of mediated Na-O2 and Ca-O2 metal-air cells, with a particular focus on Na-O2 discharge mediation and the ability of mediators to tune different peroxide/oxide deposition pathways in Ca-O2 cells. 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 award will support research towards the development of millimeter-scale flying robots capable of controlled operation without a physical connection for power or data. The robots will be powered and controlled remotely by a single-axis alternating magnetic field. Careful design of the flying structures provides passive aerodynamic and gyroscopic stability. This project will focus on providing extended range and enhanced maneuverability toward successful deployment in various applications. This project draws upon disciplines including aerodynamics, control, microelectromechanical systems, and 3D printing to advance the state of the art in remote power transmission and flight control for small rotorcraft. Results from this research will benefit the US economy and society by significantly reducing the size and weight of untethered flying robots. The multi-disciplinary approach will help increase participation and positively impact engineering education. This project plans two key technical innovations for millimeter-scale, untethered, remotely powered flying rotorcraft. The first innovation is utilizing the gradient of magnetic fields to regulate lateral flight motions. The second innovation is beamforming of the magnetic field for extending the range of operations. Flying insects have been the inspiration of miniature flying robots toward innovations and breakthroughs that are potentially applicable beyond the field of robotics. However, the tradeoff between mass and power becomes problematic for engineered systems at these scales. Specifically, the low energy densities of current options such as batteries or supercapacitors make them impractical for on-board power storage. Oscillating magnetic fields have been demonstrated to remotely power millimeter-scale rotorcraft. Because magnetic field strength attenuates with distance following an inverse square law, a simple point source will not deliver sufficient power to a flying robot at a practical operating distance. The purpose of the beamforming task is to use an array of emitters with controllable phases and amplitudes to project the field only to points where it is needed. To carry out useful functions the robots must be steerable. To this end, the research team will investigate flight control and maneuverability through control of the magnetic field gradient. 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

Genome engineering holds great promise for producing new plant varieties better able to feed the increasing global population. In many species, this approach requires selectively coaxing small clumps of cells carrying DNA sequences of interest to regenerate into whole plants with well-organized shoots and roots. Although regeneration is readily achieved and allows for successful genome engineering in some plant species, regeneration is still too low-throughput or prohibitively difficult in many crops. A critical challenge then is defining and overcoming the barriers that frustrate regeneration in these species. The proposed research will address this problem by investigating why whole plant regeneration is so hard to achieve in sunflower, a globally significant oil and confectionary seed crop. Specifically, the responses of cultivated sunflower cells and tissues to various experimental regeneration conditions will be characterized, and these observations will be compared to the responses of more easily regenerated relatives in the sunflower family. These results will then be leveraged to design experimental interventions with the goal of overcoming sunflower’s barriers to regeneration. Thus, the proposed research will deliver a toolkit of solutions for a major crop that currently lacks modern biotechnology tools to validate gene functions and accelerate breeding for traits essential to sunflower biology and agricultural production. The findings and methods from the proposed research will be communicated to the sunflower research and plant biotechnology communities through a workshop and online materials. Multiple early-career researchers will receive cross-disciplinary training over the course of the project. Recent successes in overcoming barriers to plant regeneration have been achieved by directly manipulating the transcription factors that promote shoot initiation and patterning. These gains have come largely in transformable species, rather than in species recalcitrant to regeneration, in which additional undetermined inhibitory mechanisms may be active. The proposed research seeks to define the gene regulatory programs that inhibit regeneration and test experimental means to overcome them in Helianthus annuus, the common sunflower. Using both tissue-level analyses and single nucleus transcriptomics, the project will characterize the meristematic cell types and gene expression patterns that arise during regeneration in the absence and presence of various regulators of shoot development. A comparative experimental approach will be taken to assess how growth and transcriptional responses to treatments with regeneration-promoting developmental regulators differ between regeneration-recalcitrant sunflower and two readily regenerable relatives in the Asteraceae family. Experiments will also be performed with both wild and domesticated sunflower accessions to test whether changes that evolved in developmental gene networks during domestication are a major contributor to cultivated sunflower’s recalcitrance to regeneration. Insights gained from identifying the molecular mechanisms that inhibit regeneration in sunflower and from additional plans for methods development will contribute to broader efforts in plant biotechnology and support transformation in other recalcitrant species. Together, the proposed research will not only deliver long sought biotechnology solutions for plant breeding in a major crop, but the findings will also support development of new techniques to engineer other species important to the bioeconomy. 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

Elementary particles interact, which means they impart forces on each other that can attract or repel. These forces determine how matter is built up from the fundamental particles, from atomic nuclei and atoms to molecules and solids. The fine-structure constant is a number that specifies the strength of electrical forces. Precision measurements of the fine-structure constant are important for many areas of physics, as they provide one of the most precise tests of the standard model of particle physics. These measurements can also be used to search for new particles. In 2018, with NSF support, the PI and coworkers published one of the most accurate measurements of the fine-structure constant to date. Subsequently, they have built a new apparatus with the goal of making an even more accurate measurement. In this new project they will complete this measurement by introducing three innovations that are intended to speed up progress and increase precision. As a broader benefit, this work will train undergraduate and graduate students in the technologies that enable quantum information science. This experiment functions by using photons to impart momentum kicks on atoms, and measuring the speed of the resulting atomic motion. From these measurements the fine-structure constant can be deduced. In this updated version of the experiment, the PI and coworkers will introduce three innovations. First, they will introduce a small temporal offset between two measurements that are usually done simultaneously. This will enable them to direct all of the laser power to just one of these measurements at a time, allowing them to use a larger laser-beam diameter. Second, they will make the paths taken by the atoms more symmetric, which will increase precision. Third, they will perform detailed modeling of the atoms’ paths to understand possible sources of measurement errors. 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

The Center for Genetically Encoded Materials (C-GEM) develops new biologically inspired methods to precisely tailor and build molecules. C-GEM is re-engineering the ribosome and its attendant enzymes to build complex molecules one bond at a time in a precise sequence, much like natural ribosomes build proteins by assembling alpha-amino acids in sequence. They are also inventing new chemistry to subsequently modify these materials to introduce additional functionality. Through the research, C-GEM trains collaborative teams of students and postdoctoral researchers at a rich new interface of chemistry and biology. C-GEM also engages citizen scientists with a scientific discovery game, EteRNA, that introduces interested gamers to the key concepts of RNA structure, stability, and function. C-GEM’s research plan includes interrelated goals that will expand ribosome-mediated chemistry to generate molecular architectures unique in both structure and function. Novel experimental and computational tools to design, evolve, and characterize the molecules of translation will expand their chemical capacity in vitro and in vivo. The novel molecules and materials produced will probe the fundamental chemistry of peptides, impact human health and open new opportunities in bio-materials design. Education and participation programs integrate research, workforce development for biotechnology, and public engagement. 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

Microbes are being used to produce a range of products. Some of those products are proteins, such as insulin or blood-clotting factors. In many cases, fungal strains are used because they have protein secretion systems. However, the secretion systems are not optimized for throughput, so strain improvement is needed. To facilitate this process, AI methods will be employed to develop a digital model that emulates most of the behaviors of the original fungal cells. This is commonly referred to as a digital twin. The digital twin will be used to identify genetic edits that will improve protein synthesis and secretion. This will be accomplished for many fungal strains and behaviors. The models will be made publicly available using open-source software. A hierarchical genome-to-phenotype model (G2PM) for fungal systems will be developed. The focus will be on Pichia strains. This model will link DNA sequence to gene expression and, ultimately, to strain-level protein secretion performance. More than 8,000 diverse fungal genomes will be curated to train genomic language models (gLMs). These are deep neural networks that learn complex probability distributions over nucleotide sequences. These pretrained fungal gLMs and their learned embeddings will then be integrated into a sequence-to-expression model that predicts high-resolution RNA-seq profiles directly from genomic sequence, across multiple fungal species and environmental contexts. In parallel, the team will generate and publicly release a comprehensive multimodal dataset of engineered Pichia strains. Each will be annotated with whole-genome sequence, transcriptome profiles, and single-cell secretion measurements. Leveraging this resource, the G2PM will combine the pretrained sequence-to-expression module with protein representations in a phenotype-prediction module to predict secretion titers for target proteins from genomic sequence alone. Together, these efforts could yield an AI framework capable of recommending genomic edits that enhance protein secretion, thereby accelerating fungal strain engineering and enabling more efficient, lower cost biomanufacturing. This project is being jointly supported by ENG/CBET/CBE, BIO/MCB/SSB, and the BioMADE Manufacturing Innovation Institute. 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 award funds the research activities of Professors Hitoshi Murayama and Lawrence J. Hall at University of California, Berkeley. Major advances are expected in understanding the laws of physics never probed before, with new data coming from the Large Hadron Collider in Europe (which is currently the highest-energy colliding beam facility in the world), from underground experiments in South Dakota, and from other laboratories around the world. Additional data is also expected from precision experiments at Fermilab (the Fermi National Accelerator Laboratory in Illinois), with additional insights coming from deepening connections to experiments in astrophysics and cosmology. Research in theory is needed to tie all of these results together into a coherent framework. In this project, Professors Murayama and Hall aim to uncover deep secrets that span the range from the smallest scales of particles and strings to the largest scales of the Universe. The scope of their research will include collider physics, dark matter, neutrinos, quark flavor, the phenomenology and theory of supersymmetry, observational cosmology, gravitational waves, strongly coupled quantum field theories, and the multiverse. It is also expected that new collaborations will emerge on these topics. Professors Murayama and Hall will also be active in public outreach and training the next generation of researchers as well as members of the general scientific workforce, thereby situating this work within the national interest. More technically, Professor Murayama will study non-perturbative dynamics of strongly coupled gauge theories, exploiting their exact solutions in the supersymmetric limit together with anomaly-mediated supersymmetry breaking. This method has allowed people in the community to test many ideas concerning non-perturbative physics, including axions, phase transitions, oblique confinement, etc. Professor Murayama will also build models of dark matter and axions covering different parameter sets relevant for experiments. Professor Hall will work on developing new models of flavor, both at the grand-unification and multi-TeV scales. These models will lead to new results for precision flavor physics experiments. Such results can also potentially address the strong CP problem without an axion. In addition, Professor Hall will incorporate the multiverse paradigm together with symmetries in order to understand the origins of flavor and neutrino mass. 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.