De Tenebris (“On Darkness”): Dark matter phenomenology in strong gravity

De Tenebris (“On Darkness”) pioneers a new way to detect dark matter - the elusive substance comprising most of the Universe’s mass - by "listening" to ripples in spacetime called gravitational waves. Bertone’s project will develop models and computer codes to predict exactly how different types of dark matter would alter a black hole merger signal, and to identify key waveform signatures that reveal dark matter’s nature. 

GT4Pebbles, Ground-Truth for Pebbles

To uncovering the origins of Earth, we can observe the disks of dust and gas of (exo)planets, the place where planets are born. Dust grains colliding and sticking together, gradually forming compact millimeter- to centimeter-sized particles, are known as Pebbles. Once Pebbles exist, we understand how full planets can form. But can enough Pebbles form with the right properties? Dominik’s project aims to answer this through advanced simulations of dust growth and compaction in planet-forming disks. 

UNITE: Robust Generative Information Retrieval

Search engines often prioritize immediate results over broader societal values like fairness and diversity. De Rijke’s project tackles this challenge head-on, proposing a paradigm shift towards generative retrieval systems. 

The Common, the Rare, and the Unseen: Precision Theory for Precision Experiments

Particle physics faces major challenges, such as the absence of antimatter and the origin of neutrino masses. While particle physics is typically associated with high-energy colliders, low-energy high-precision experiments offer a promising alternative for making new discoveries. The success of the precision programme depends on a combination of both theoretical and experimental accuracy. In his project, De Vries will develop new tools to enhance the accuracy of key measurements of well-known particle decays (the common), extremely infrequent atomic processes (the rare), and yet-to-be-discovered phenomena (the unseen).

Cell Type-Specific Disease Resistance Against Vascular Xanthomonas Plant Pathogens (XANTHORESIST)

Plants have fascinating ways to protect themselves from harmful germs. They use physical barriers to keep invaders out and an immune system to fight off infections. But germs have their own tricks to get inside plants and cause disease. Stopping germs at the water pores could prevent a systemic infection. As we currently know very little about how plants defend this entry point, Pfeilmeier’s project aims to identify the defence responses in the plant cells that first perceive and fight the harmful invaders. 

Unraveling the nature of fast radio bursts through a multifaceted look at their local environments (EnviroFlash)

Fast radio bursts (FRBs) originate far beyond our Milky Way and were only recently discovered. It is still a mystery where and how these bursts are produced, but it is clear that some sources repeat while others apparently do not. The key to understanding the variety of FRBs lies in their local environments. In his project, Pleunis will capitalise on the ‘Outriggers’ upgrade to the CHIME telescope, which has transformed the world’s best FRB detector into a machine that can now also pinpoint the bursts to their galactic neighbourhoods.

A Neonatal anaerobic Gut-microbiome-on-a-Chip to decode bacterial colonization and infant gut T cell maturation (NeoGutChip)

The early days of life are a critical window to develop a healthy relationship with the infant gut microbiota. Because these early residents can train our naïve immune system for later challenges such as infections and allergens. Despite its importance, our understanding on this training process is limited. Therefore, there is a pressing need for an infant gut model to capture the training of beneficial microorganisms on our immune system in early life. Jianbo Zhang will develop a first-of-its-kind avatar of the human infant gut, an in vitro infant Gut-Microbiome-Immune-on-a-chip model.

Asymptotic spectra: from algebraic complexity theory to graph theory and beyond (SPECTRA)

When we repeat a task many times, how can we do it most efficiently? This simple question lies at the heart of many challenges in computer science, mathematics and physics: from finding faster computer algorithms, to determining the limits of data transmission in networks, to the optimal packing of geometric shapes. In his research project, Zuiddam will develop a mathematical framework (the asymptotic spectrum) to tackle such ‘economies-of-scale problems’.

INtelligent System for Processing Intensive Reaction Electrochemistry (INSPIRE)

One of the challenges in synthetic chemistry is making the large-scale production of pharmaceuticals and bioactive molecules more sustainable. Electrochemistry offers a potential solution here, but optimizing electrochemical synthesis routes remains a significant challenge. Hopsort will tackle this by developing a fully automated system that couples continuous-flow electrochemistry with artificial intelligence for optimal process development.

Dr Salvatore Romano will work for two years on a sustainable bio-organocatalytic cascade for amide synthesis from carbonyl compounds. Designed with green chemistry principles, the method will provide an efficient, scalable solution for pharmaceutical manufacturing of high-value compounds, including chiral amides.

Synthetic robots for safer and greener chemistry

Chemistry labs have long relied on manual processes that are time-consuming, expensive, and sometimes hazardous. SynthBot is a groundbreaking robotic platform that will automate key steps in chemical synthesis, from discovering new reactions to optimizing them. By using machine learning and 3D-printed components, SynthBot will make chemical research faster, safer, and more efficient. This innovation has the potential to reduce waste, improve safety, and make advanced chemical techniques more accessible to researchers in academia and industry, ultimately paving the way for new breakthroughs in medicine, materials, and sustainability.

How the sloppy table manners of black holes and neutron stars shape our universe

Black holes and neutron stars are notorious for their enormous gravity that allows them to devour anything near them. However, these cosmic cannibals are messy eaters and fling gas and energy into space via so-called jets. These carry such enormous power that they play a key role in our universe. For instance, jets determine how galaxies develop, produce ultra-fast particles and enrich the cosmos with exotic elements. How jets are launched and energized remains a mystery. In this project, an innovative technique is applied to astronomical observations of large telescopes to enforce a breakthrough in understanding jets.

Mathematical foundations for explainable Artificial Intelligence

Artificial intelligence (AI) keeps getting better. How it makes its decisions often remains unclear, because these depend on many incomprehensible numbers in the software. In this project researchers will develop new mathematical techniques that can explain AI's internal mechanisms by generating understandable explanations from these numbers. They will achieve this by creating the first mathematical framework for explainable AI and applying advanced methods from mathematical machine learning theory.

BrainShorts: Identifying neural mechanisms of dynamic video perception with deep learning

Our visual brain has no trouble extracting meaning from a short video, while advanced AI techniques require a lot of computing power for the same feat. What calculations and clever tricks does the brain use to achieve this? Groen’s project will answer this question by first collecting high-resolution brain measurements from the video-watching brain and then mimicking these with deep learning models that we enrich with various biological information processing principles. By directly comparing models with and without brain-inspired computations, Groen will identify key biological processing principles, facilitating more energy-efficient AI.

Identifying neural mechanisms of dynamic video perception with deep learning

Our visual brain has no trouble extracting meaning from a short video, while advanced AI techniques require a lot of computing power for the same feat. What calculations and clever tricks does the brain use to achieve this? Groen’s project will answer this question by first collecting high-resolution brain measurements from the video-watching brain and then mimicking these with deep learning models that we enrich with various biological information processing principles. By directly comparing models with and without brain-inspired computations, Groen will identify key biological processing principles, facilitating more energy-efficient AI.

HUMBOLDT—High resolUtion Mountain Building of the COLombian AnDes through Time

Mountains are known as biodiversity hotspots, yet what drives the extraordinary diversity of many mountain ranges remains a profound mystery. Ott’s project will explore whether the—geologically speaking—rapid rise of mountain ranges acts as a catalyst for increased biodiversity. This understanding is vital for establishing conservation priorities. Ott’s project will focus on the Northern Andes—the world’s most biodiverse mountain range—and collects data to reconstruct their rise over time. Ott will analyse how the speed and variability of mountain formation impact biodiversity. This will reveal how dynamic geological processes influence the evolution and abundance of life.

CANES: a CAusal NEuro-Symbolic approach to integrating perception and abstract reasoning

Current AI systems often make decisions in ways that people don’t understand. This potentially creates mistrust, unfairness and brittleness of these systems to changes. A solution is to build systems that learn and use understandable concepts. This requires annotators to label large amounts of data, which is tedious and expensive. Worse, even with all these annotations, AI models can still learn the wrong concepts! Magliacane will develop new methods that learn understandable concepts correctly with a high probability from few annotations. She will also use available knowledge on the interactions between the concepts and information on past decisions.

iGuMI: An infant Gut-Microbe-Immune-on-a-chip model to decipher crosstalk between bacterial colonisation, mucosal barrier integrity and macrophage function in early life

The first 1000 days of life are crucial for microbiome development, immune system formation and long-term health. However, yet we still lack a full understanding of how intestinal bacteria shape immunity. Zhang’s project aims to develop a novel miniature gut-microbe-immune-on-a-chip (iGuMI) model that mimics the infant gut to unravel how "good" andc"bad" bacterial colonisation influence infant intestinal function and immune development, identify key bacterial species and metabolites that drive these effects. Ultimately, this research will inform the development of new interventions (probiotics or drugs) to boost infant immune health, prevent infections, and improve pediatric medicine.

Asymptotic spectra in mathematics and computation

What is the cost of a task when performed many times? This question lies at the core of central problems in mathematics, computer science, and quantum information—such as fast matrix multiplication, Shannon capacity, and quantum entanglement transformations. Zuiddam’s project will advance the theory of asymptotic spectra, a powerful new approach that reveals deep structure in these problems across disciplines. By developing this theory and applying it to direct-sum problems, Zuiddam aims to overcome long-standing barriers through novel methods from algebra, topology, combinatorics, and optimisation.

HyperVision: Hyperbolic Computer Vision

Neural networks excel at recognising what happens in images but suffer from a crucial blind spot: they fail to correctly represent hierarchies. Hierarchies are key structures in computer vision, describing how visual data and semantics are organised. The mismatch with hierarchies has a geometric reason: modern networks are Euclidean, while hierarchies have a hyperbolic nature. To overcome this fundamental problem, Mettes’ project proposes to redefine and rebuild neural networks for computer vision using hyperbolic geometry, with the promise of addressing critical limitations of today’s neural networks.

Cell Type-Specific Disease Resistance Against Vascular Xanthomonas Plant Pathogens (XANTHORESIST)

Plants have an immune system that detects and responds to pathogens, yet many crops are susceptible to infections. Cabbage and related vegetables, for instance, are particularly vulnerable to Xanthomonas bacteria. These bacteria enter through leaf water pores and spread through the plant’s vasculature, causing Black Rot disease. The mechanisms by which plant cells recognise Xanthomonas and initiate defence responses remain unclear. Pfeilmeir’s project aims to identify key immune processes during the early stages of infection, providing new insights that can be applied to develop disease resistant crops.

Radio Stars and Exoplanets: Creating space weather reports for distant worlds

Understanding whether exoplanets can support life is a fundamental goal in astronomy. Space weather—such as magnetic fields and plasma eruptions from stars—plays a crucial role in determining whether planets can retain their atmospheres under the influence of stellar activity. Callingham's project focuses on detecting exoplanets and plasma eruptions from stars through their radio emissions. He will use this information to create the first space weather reports for other worlds—an essential step in the complex puzzle of finding habitable planets.

ROOTS: Resilient outcomes from organic transitions in society

Vasconcelos’s project explores how people can be supported in making more sustainable choices—like installing solar panels or eating less meat—without fuelling social division. To understand how opinions and behaviours shift, he will collect and analyze large datasets from across Europe and the US. These insights will feed into computer models that test which climate policies are most effective, and under what conditions they might unintentionally increase tensions between groups. Vasconcelos’s project will work closely with societal partners to ensure that climate measures not only support sustainability but also strengthen social cohesion.

The Hunt for New Physics: Using the Power of Neutron Stars to Test Einstein's Theory of Gravity 

Brown's project investigates binary neutron star mergers, exploring the intersection of gravitational and nuclear physics. These extreme events, with strong gravitational fields and near-light-speed dynamics, provide a unique opportunity to study fundamental physics and the nature of ultra-dense matter. By developing a framework to test general relativity that accounts for neutron star properties, Brown will enable the first theory-agnostic analysis of such systems. Additionally, the project will advance data analysis and machine learning techniques, maximizing the scientific potential of current and future gravitational wave detectors and opening new pathways to understand gravity and the universe.

Human-Centered Code: Building Accessible IDEs for Neurodiverse Women in Computing Education

Brooke's project addresses the lack of diversity in software engineering by focusing on neurodiverse women learning programming. Using generative AI, Brooke will create personalised, inclusive tools to support diverse learning needs. Through focus groups, co-design workshops, and iterative design, the project identifies barriers in existing coding tools and develops AI-powered interventions. It introduces the “human-centered code” framework, embedding user-driven, inclusive principles into programming environments. By making accessibility and equity central to coding tools, this research empowers neurodiverse women, enhances inclusivity in programming education, and positions AI as a force for collaboration and innovation.

Chemistry in Ionospheres as a New Way to Probe Atmospheric Conditions of Exoplanets

Planets beyond our Solar System can have atmospheres. But which molecules these are composed of is not well known. Nonetheless, the atmospheric composition reveals how a planet was formed and whether it may host life. Baeyens' project will use the James Webb Space Telescope is used to observe the atmospheres of exoplanets, and show us how the atmospheric composition can be changed by starlight.

Modular Forms and Multiple Zeta Values in Mathematical Physics

Not only mechanics need tools: theoretical mathematical research also requires utensils—abstract hammers and saws, that is. Van Ittersum's project develops all kinds of tools (mathematical theorems) to describe sequences of numbers in number theory, geometry and mathematical physics. In particular, van Ittersum will determine the ‘fingerprint’ of new so-called modular symmetries that such sequences can possess, with the aim of making such symmetries easier to recognize and use. In doing so, the project will contribute to the significant collection of mathematical problems solved using modularity properties.

From Theory to Test and Back: Testing and Advancing Theory about Sensitive Periods beyond Early Ontogeny

An evolutionary enigma is why individuals are so sensitive to experiences at certain life stages. Most research focuses on these sensitive periods early in development. But there is growing awareness that later stages, such as adolescence, have heightened sensitivity too. However, formal mathematical models or empirical tests on the evolution of these later-life sensitive periods are lacking—yet important for predicting how animal populations will respond to rapid climate change or major adverse events (e.g., a global pandemic). Through a powerful combination of experimental evolution and novel theoretical models, Walasek's project will advance our understanding of later sensitive periods.

Foundations of Sum-of-Squares Algorithms: Worst-Case and Beyond

Sums of squares are mathematical objects with a long history. Historically, they have been studied mostly because of their theoretical properties. More recently, they are have been used as a tool in the development of powerful algorithms. These algorithms can be used to solve all sorts of problems in pure and applied science. In this project, Slot will expand the mathematical foundations of these algorithms in an attempt to explain their success. In particular, he will develop methods that allow us to go ‘beyond the worst case’.

This project at the UvA develops eco-friendly borate electrolytes for sodium-ion batteries (SIBs) using borax and industrial boron waste. It reduces environmental impact and reliance on scarce materials while offering a safer, more sustainable alternative to lithium-based systems (LIBs). The goal is to create affordable, reliable energy storage for renewable sources.

Microbes for a Sustainable Agriculture

Agriculture is crucial for providing food to a growing world population, but faces large challenges. Many of these can probably be overcome with microbes. The start-up company MicroBioLogicals will develop a product against potato cyst nematodes based on microbes. After this first product, MicroBioLogicals aims to develop more products, also based on microbes, to make agriculture more sustainable.

How nanoplastics change proteins

Plastic nanoparticles are everywhere, including in our own bodies, and there is increasing evidence that these nanoparticles affect our health. The biological effects of nanoparticles are ultimately due to their interaction with proteins, the molecules that catalyze all reactions in the cell. However, whereas intensive research is being done on the biological effects of plastic nanaparticles, very little is known about the underlying nanoplastic-protein interactions that give reason to these biological effects. In this project, Woutersen uses advanced spectroscopy methods to investigate how plastic nanoparticles change the structure and aggregation behavior of proteins.

Tumor attacking behavior of T cells under scrutiny

T cells have a unique capacity to recognize and kill tumour cells. However, we still know very little about their specific behaviour that leads to effective elimination of cancer. By combining expertise in microscopy (to follow T cells in space and time during tumour attack) and computer vision (to recognize behavioural patterns in the resulting microscopy datasets), Dr Anne Rios (PMC) and Dr Efstratios Gavves (IvI) hope to gain a better understanding of the behavioural strategy of T cells leading to elimination of tumour cells.

Precision matters

We, the earth, the stars – everything in the universe is made of matter and not antimatter, and we do not know why. An exciting possibility is that this matter-antimatter asymmetry was created just picoseconds after the Big Bang, when the matter particles became massive. This research proposal aims to very precisely alculate the predicitons for this scenario, and to confront it with the wealth of current and upcoming experimental data. 

The effect of stress hormones on memory-encoding brain cells

Stressful experiences are generally remembered well, but such memories are often less precise, which results in memory generalization (recall of the stressful event when this is not relevant). This effect is mediated by stress-hormones. Dr Harm Krugers (SILS) and Dr Michel van den Oever (VU) will investigate how stress-hormones enhance memory generalization by studying the properties of the specific cells in the brain that store a memory. 

More powerful proofs and programs with geometry

Types theories are systems for proving mathematical theorems and writing computer programs. For this reason type theories provide the basis for software for verifying the correctness to develop a more expressive and powerful type theory. To show that this new type theory still has the beautiful computational properties that the implementations rely on, this proposal will develop a better computational understanding of important geometric ideas.

Reconstructing 25% of the Universe with the Large Hadron Collider

The Large Hadron Collider is much more than a Higgs-particle discovery machine. It has recently evolved into a dark matter discovery machine, with- upgraded data selection systems called "triggers". These triggers allow physicists to detect new particles that escape all previous experiments. We take this unique opportunity to search for dark matter candidates, which look like atoms and can leave peculiar traces in the detectors. If we observe such a dark matter atom, we will be close to unravelling the mystery behind all the invisible matter in the universe.

Economies of scale tensors

The possibility of achieving cost savings when performing many instances of a task in a large batch, as opposed to performing each seperately, is of fundamental interest in applications in mathematics, physics, economics, and computer science. Many such applications (e.g. in algebraic complexity theory, quantum information theory and combinatorics) are modelled by tensors and their asymptotic ranks. This project develops and applies variuos novel methods to make progress on these problems and gain new understanding of the asymptotic behaviour of tensors.

Guaranteeing Fairness when Everyone is Selfish

Reiffenhäuser will research how to divide resources between different parties fairly, even if they don’t tell us the truth about which ones are most important to them. With this, inheritance, divorce, kindergarden days and all kinds of other things where participants would do anything to increase their own gain can finally be settled without worrying that the result will end up unfair.

Molecular Nanoconfinement in Photochemical Charge Generation and Transfer

Properties of molecules are strongly affected by their environment, especially solvent. Because solvent is practically absent in the solid state, its use to control the transfer of electrons in modern optoelectronic devices, such as organic solar cells, is limited. This project engineers pockets with nanometer size inside solid-state organic materials by chemical synthesis and fills them with solvent molecules. It then explores how addition of the solvent impacts the ability of the organic material to convert sunlight into charge. This research provides a better understanding of fundamental processes in organic optoelectronic devices and takes a step forward towards their innovation.

Ogg’s conjectures over function fields

Drinfeld modules are objects studied in arithmetic geometry, that may be viewed as function field analogues of elliptic curves. The behaviour of their rational torsion points is governed by (generalised) Ogg’s conjectures. This proposal addresses several Ogg’s conjectures by a new, geometric approach that involves rational points on Drinfeld modules curves.

NOMAD – Nanocarrier Optimization of Metal-organic cages for Alzheimer’s Drug delivery

Alzheimer’s – the most common neurodegenerative disorder – is a horrible disease. Current therapies are only able to reduce symptoms, not stop disease progression, let alone prevent it. Through gene therapy, we might be able to halt progress or even cure neurodegenerative disorders. However, current methods to delivery gene therapeutics do not suffice. In this project, we explore a new gene therapeutics delivery agent – metal organic cages. Through computer simulations and experimental techniques, we gain insight into cage-therapeutic binding and interactions with various neural cells that will result in optimized cages as delivery agents for Alzheimer’s treatment.

How fast do soils get lost to erosion after deforestation? 

Soils need to be used in a sustainable way, because we rely on food from agriculture. However, deforestation and common farming practices lead to unsustainable loss of soils, which reduces crop yields and releases CO2 to the atmosphere, accelerating climate change. Thus, we need to measure how fast soil is lost due to deforestation and farming practices to guide mitigation strategies. This project explores a new method for measuring erosion by combining the measurements of several rare isotopes. If successful, this method will allow to measure erosion due to human activities and climate change at spatial and temporal unprecedented resolution.

Natural Nanobatteries

This project combines two powerful tools to create natural nanobatteries. Montmorillonite clay, with its ability to swell and contract with hydration, could be the key to ultra-small, sustainable batteries. By embedding charge-generating crystals within the clay, its mechanical movement can be converted into electricity—creating a self-charging nanobattery. The challenge? Ensuring proper crystallisation and allowing for both the generation and conduction of charge. The plan? Tune the clay’s properties and the crystal size using organic molecules to control the swelling and contraction and maximise energy output. This research could pave the way for the next generation of green batteries.

Beyond the Black Box: Unlocking the Superpowers of Enzyme@MOF Systems

Enzymes are nature’s superheroes: fast, selective, and efficient. Yet many of the reactions needed for producing medicines or green chemicals, take place under harsher conditions where enzymes quickly fail. Metal–Organic Frameworks (MOFs) act as protective suits, that can extend their lifetimes for industrial chemical production. But once inside these suits, enzymes fight hidden battles in a black box, and scientists don’t know if their powers remain intact. This project uses a one-of-a-kind “X-ray vision” platform to watch enzymes in action. By revealing how protection affects performance, we aim to design better biocatalysts. 

Turn off the light to see

Dark matter makes up most of the universe, yet we cannot see it directly; it is known only by its effect on galaxies. We build sensitive detectors to catch faint signals from dark matter particles passing through Earth, but these detectors often pick up false signals instead. We believe we have found the cause of these false signals. In this project, we will build a false-signal screening station for materials used in detectors. If it works, this will both confirm our idea and help us prevent fake signals - a key step toward finally discovering dark matter. 

Photoactivated ferroptosis for cancer therapy

Targeted cancer chemotherapies use antibody–drug conjugates to guide the drug directly to the tumor. Despite substantial improvement of such targeting, off-site toxicity often persists because biochemical processes sever the antibody–drug link prematurely. This decreases efficacy and the quality of life of patients. An alternative strategy that hijacks photons and iron to induce cancer cells death avoiding toxic spillovers is proposed and validated here. The latter is done by performing the experiments on cancer cells in vitro. Thereby, the potential of this ground-breaking approach as a competitor to the current principles of chemotherapy will be explored.

Leveraging Nighttime Light Data to Reveal Global Economic Dynamics

Economic networks are highly interconnected and rapidly evolving due to globalization and technological advancements. Traditional economic analyses often face challenges such as fragmented data and difficulty distinguishing causal relationships. This project tackles these challenges by developing a temporal causal network model leveraging high-resolution nighttime light data. The research aims to uncover complex interactions within global economic networks, identifying influential clusters and revealing core-periphery structures. These insights will empower policymakers with predictive tools to guide interventions and strengthen economic resilience.

Unveiling the dark Universe with Gravitational Waves

When two black holes spiral around each other, they emit gravitational waves that we detect here on Earth. When black holes interact with dark matter, the signal is modified. Tomaselli’s research aims at characterizing this modification, with the goal to search for the “dark universe” using gravitational waves. Tomaselli will use the grant to continue his research on black holes and gravitational waves as a postdoc at the Institute for Advanced Study in Princeton.

Active Colloidal Solids: From Spontaneous Collectives to Distributed Machines

Veenstra's project develops active colloidal solids: mechanically linked assemblies of self-propelled microparticles whose elastic and active interactions generate shape change, motion, and mechanical response. Inspired by the distributed, force-coupled nature of living systems, this approach uncovers design principles for collective mechanical behavior, transforming fragile active matter into stable, adaptive materials functioning as distributed microscale machines. Veenstra will go to ENS Lyon, France, for two years.

Coursing through the Galaxy: properties of the fastest explosive stars

Massive stars will end their life in an explosive supernova. A substantial fraction of these massive stars is observed to have an extraordinarily high velocity: so-called runaway stars. Stoop's project makes use of the ESA Gaia telescope to find thousands of runaway stars and investigate their yet unknown kinematic properties. Stoop will go to Max Planck Institute for Astronomy, Germany, for two years.

Sequential deformations in pressurized shells

Pressurized shell structures are stiff and lightweight but susceptible to catastrophic collapse due to mechanical instability. By integrating the concept of sequential deformation, the researcher aims to develop novel shell designs that undergo controlled, stepwise deformation. This approach offers a promising solution to address both structural stability and functional adaptability. Liu will go to Harvard University, US, for two years.