UNIQplus projects

As part of your UNIQplus Research Internship, you will be working on a project under the supervision of academic staff from our community of world-leading researchers.

UNIQplus projects . Credit: Phil Brooks

Admission status

Closed to applications for entry in 2026.

Projects available for entry in 2026

Expand all

Mathematical and Physical Sciences

Expand all

Projects in Mathematical and Physical Sciences are offered by the following departments:

Chemistry 01
Interrogating immune signalling megastructures for new knowledge advances & drug discovery

Primary supervisor

Professor Yimon Aye

Project description

The goal of this project is to create chemical biology approaches to localize specific proteins to defined biological megastructures. We have reliable methods to create specific biological megastructures on demand through triggering 'adaptive immune cells', namely, 'T/B-cell receptor signalling'. These megastructures are hubs of negative regulation of this crucial proinflammatory pathway. We have also shown that these megastructures tolerate fusion of small self-labelling proteins to their component proteins. By designing and synthesizing appropriate bifunctional ligands, we will force proteins typically excluded from these crucial signalling complexes, to associate with them.

Once this association is validated by imaging, we will assay effects on signalling by antibody-based quantitative methods such as ELISA for specific cytokines. This will form the bedrock of a 'proteome-wide' screen for regulators of specific signalling processes. This project will teach students advanced molecular cloning, bioengineering, and chemical biology. We will also touch on immunology and drug design. All basics are in place in house.

Project outcome

You will have the opportunity to learn modern technologies and concepts underpinning biotechnology, bioengineering, gene cloning, chemical biology, and drug design. You will also have the opportunity to present your results in a group meeting, and as time/interest permits, to gain additional pedagogical experience through contributing to a short perspective/review article for publication as co-author.

Entry requirements

You should have, or be studying, a degree in Chemistry, Biochemistry, Molecular Biology, or a genetics-related degree.

Chemistry 02
New light emitting organic molecules with interesting spin properties

Primary supervisor

Dr Daniel Congrave

Project description

Organic molecules and materials that absorb and emit light broadly overarch traditional scientific disciplines with implications in applications related to energy, medicine, electronics and sensing. This project will expose you to the world-class synthesis and Electron Spin Resonance (ESR) facilities at Oxford Chemistry.

We have recently been pioneering organic molecules that can emit light while also featuring high-spin states. This is extremely promising for developing a new generation of molecules that can function as quantum sensors, with the ability to read-out information completely non-invasively through fluorescence.

You we be involved in developing our new fluorescent high-spin molecules, having the opportunity to see how physical and organic chemistry can link together and to learn a wide range of skills associated with this interdisciplinary research.

Project outcome

This project is interdisciplinary and combines organic and physical chemistry. It is therefore an ideal opportunity to learn a wide range of skills and be broadly exposed to scientific research so that you can make an informed choice in what to pursue in your research career after the internship. You will be involved in a collaboration between organic and physical chemists, which will provide an opportunity to improve your scientific communication skills.

You will have the opportunity to take a never before synthesized molecule from the conceptual design to the final characterization experiments after synthesizing it. You will firstly be made familiar with the computational chemistry used to design new target molecules. You will next be trained in organic synthesis, particularly air- and water-free experimental techniques, and be exposed to multiple methods of chemical purification such as chromatography and distillation. Once the new molecule is synthesized you will have the opportunity to take part in characterization experiments such as NMR, absorption and fluorescence spectroscopy, and ESR, and then be taught how to interpret the data.

You will be directly supervised in the synthetic organic chemistry lab by a PI who has a decade of experience in this type of research, and will be able to observe and be exposed to the wide range of other organic optoelectronics projects being carried out within the group. At the end of the project you will have the opportunity to write a report and may present your results to collaborative team. If any aspect of your work is included in a future publication, you may be included as a named co-author on that paper.

Entry requirements

You should have, or be studying, a degree related to Synthetic Chemistry (eg Chemistry or Natural Sciences). Some familiarity with synthetic chemistry lab work typical of that encountered in an undergraduate degree is desirable.

Chemistry 03
Decoding the light of fungi: Exploring the role of hispidin 3-hydroxylase in bioluminescence

Primary supervisor

Dr Patrick Rabe

Project description

Hispidin 3-hydroxylase (H3H) is a key enzyme in the fungal bioluminescence pathway of Neonothopanus nambi, catalysing the final step in luciferin biosynthesis required for light emission using molecular oxygen. This self-sustaining pathway enables continuous bioluminescence, making it an attractive tool for biomedical imaging and the development of autonomous bioluminescent plants.

The project will focus on the biochemical and structural characterisation of H3H and selected mutants to define the enzyme’s active site and elucidate its reaction mechanism, identifying structural features or mutations that could optimise the wavelength and intensity of light emission. This will be achieved using advanced chemical biology approaches, biochemical assays, and recombinant protein expression, alongside crystallographic and biophysical methods to provide a detailed structural and functional understanding.

Ultimately, this knowledge will support the bioengineering of the bioluminescent pathway into plants for sustainable, tuneable carbon-neutral lighting and into other cellular systems for drug discovery and imaging.

Project outcome

The Avestro Group is interested in escaping this popular 'flatland' paradigm by instead exploring the potential of 3D conjugated molecular shapes such as molecular hoops, spirals, tubes and frameworks to promote good electron sharing and energy storage across precisely positioned electron-storing sites in the structure. What are the advantages and added benefits of introducing this additional structural dimension to organic energy storage materials? This project aims to find out!

Project outcome

You will have the opportunity to synthesise new organic molecules that are capable of accepting and releasing charge on demand, and sharing electrons over 'long' distances on the molecular scale, which make them attractive systems for studying fundamental electronic properties that could reveal new design principles for functional materials. You will next be trained in air- and water-free organic synthetic techniques, including the use of microwave reactors to facilitate reaction times and encourage environmentally friendly conditions. You will also gain hands-on experience in cutting edge chemical purification such as automated chromatography, which will give you exposure to methods employed by chemical industry. Once new molecules are synthesized you will have the opportunity to take part in characterization experiments such as NMR spectroscopy, absorption/fluorescence and electrochemistry of their charge-released and charge-storing states -- and of course, be taught how to interpret the data! -- to understand how the charge carriers (electrons) behave within the molecular structure and unravel fundamental structure-property insights.

You will be mentored directly by the supervisor and an experienced DPhil student, participate in weekly informal research meetings to discuss current activities and goal planning for the week, and gain writing technical writing experience through the delivery of monthly reports. Summer is a typical time for the Avestro Group to also have their summer team building and ideas retreat, which we would hope you can join and bond even more with the team. At the end of the project, you will have the opportunity to write a report and may present your results to collaborative team. If any aspect of your work is included in a future publication, you may be included as a named co-author on that paper.

Entry requirements

You should have, or be studying, a Master's (ie four-year) degree in Chemistry or a related chemical sciences subject. An interest and demonstration of strong organic synthetic skills is advantageous, as is an interest in analytical techniques applied to study the structural dynamics, electronic and optical properties of organic molecules (eg NMR spectroscopy, absorption and/or emission spectroscopy, electrochemistry). You will be enthusiastic and motivated with the ability to work independently where required.

Chemistry 07
Synthesis of aryl-fused bicycloheptanes (aryl BCHeps) as naphthelene bioisosteres.

Primary supervisor

Professor Angela Russell

Project description

While naphthalene and equivalent heterobiaryl rings are often encountered in drugs, they can be susceptible to metabolism, exhibit flat, sp2-rich structures and are almost invariably derived from fossil fuel sources, limiting their current and future utility. Yet, few viable alternatives have been described.

In this project we propose to build on our preliminary findings that aryl-fused bicyclo[3.1.1]heptanes (aryl BCHeps) can serve as effective bioisosteric replacements for naphthalene or (iso) quinolines, to optimise the synthetic route to improve efficiency and scope. The project will blend organic synthesis with computational chemistry, with an opportunity for some biochemistry. This project is linked with the EPSRC CDT in Chemical Synthesis for a Healthy Planet.

Project outcome

You will gain skills in methods and analytical techniques in organic synthesis and modern photocatalysis (Angela Russell lab), and also gain insights into and experience of computational analysis to improve reaction design (Fernanda Duarte lab). The project can encompass the design and biological testing of benzoBCHep as arylhydrocarbon receptor antagonists, giving also practical experience in biochemistry.

Entry requirements

Chemistry 08
Improving the sustainability of small ring bioisostere synthesis

Primary supervisor

Professor Edward Anderson

Project description

Bridged small ring scaffolds are becoming increasingly important in medicinal chemistry research as novel building blocks in drug design that exhibit superior properties compared to traditional motifs such as benzene rings. However, methods to access these important molecules have typically relied on pyrophoric / hazardous reagents such as organolithiums.

In this project we address this challenge by developing novel approaches to cage frameworks that do not rely on such conditions. The research will involve an array of organic synthetic chemistry methods and analytical techniques. This project is linked with the EPSRC CDT in Chemical Synthesis for a Healthy Planet.

Project outcome

You will receive training in organic synthesis and in NMR analysis of small ring frameworks, including stereochemical and regiochemical assignment. You will be required to research the project via literature searches and online databases, and will present your findings in informal fortnightly group meetings. You will summarise your findings in a report, providing training in scientific writing.
Entry requirements

Earth Sciences 01
Analysing Earth’s ancient magnetic field using some of Britain’s oldest rocks.

Primary supervisor

Dr Claire Nichols

Project description

You will analyse the palaeomagnetic signals recorded in some of Britain’s oldest rocks (metamorphic rocks, the Corodale Gneiss, from the Lewisian Gneiss Complex in South Uist) to help understand what Earth’s magnetic field was like around 1.5 billion years ago. These rocks have not previously had their magnetic properties studied; the student will generate a new dataset that will compliment published data on the Outer Hebrides. The work will be a valuable contribution to existing palaeomagnetic datasets as there is limited information about the strength of Earth’s magnetic field around 1.5 billion years ago.

You will conduct palaeomagnetic experiments (eg demagnetisation and intensity experiments) combining these with petrography (eg optical microscopy) with the potential to also conduct quantitative mineral composition analyses (eg using SEM/EPMA) to determine the magnetic and metamorphic history of rock samples of Corodale Gneiss from South Uist, Outer Hebrides, Scotland.

Project outcome

You will gain experience in working in a laboratory setting, particularly the skills associated with conducting palaeomagnetic experiments. You will develop skills in microscopy applied to metamorphic rocks and mineralogy. You will write a short report documenting your findings, with assistance from the PI’s. The work that you will do will contribute to ongoing work the supervisory team is conducting on the tectonic and magnetic history of northwest Scotland and the Outer Hebrides, that may have broader implications for our understanding of the evolution of Earth’s magnetic field in the deep past and Archean and Proterozoic tectonics. You will be listed as an author on publications from the supervisory team that your work has contributed to. You will have the opportunity to meet with other members of the PI’s research groups and practise science communication by virtue of informal presentation of your work to peers.

Entry requirements

You should be have, or be studying, a STEM-related degree. A passion for geology/earth and planetary science, or magnetism would be beneficial.

Earth Sciences 02
Tracking ocean waves as they hit the seafloor: a seismology-based analysis

Project description

This project investigates how large language model (LLM) agents can communicate efficiently with both humans and other agents. As multi-agent systems become more common, where AI agents collaborate, negotiate, or share information, communication efficiency becomes critical for performance and safety.

The project aims to design and analyse methods that reduce ambiguity and resource usage in these interactions, such as adaptive prompting, message compression, and causal analysis of communication failures. You will gain hands-on experience with modern LLMs (eg GPT, LLaMA) and simulation environments where multiple agents interact.

In collaboration with Microsoft, the project bridges natural language processing, causal reasoning, and distributed AI to develop agents capable of clear, efficient, and robust communication, essential for next-generation AI systems that reason and act together.

Project outcome

You will gain hands-on experience working with large language models (LLMs) such as GPT and LLaMA, learning how to design and evaluate communication strategies between humans and AI agents, as well as between multiple AI agents collaborating on shared tasks. You will be trained in prompt engineering, conversational data analysis, and basic causal reasoning techniques to diagnose communication failures and improve efficiency. The project will also provide exposure to tools for running and analysing agent interactions. You will work closely with researchers from Oxford and Microsoft, gaining insight into how research ideas are translated into real-world applications.

Entry requirements

You should have, or be studying, a Computer Science related degree and have standard software development skills at an undergraduate level. Familiarity with Python and machine learning libraries would be useful. Understanding and had previous projects in LLM and agents is preferred.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Engineering 03
Improvement and commissioning of a plasma torch facility for high-temperature material testing

Primary supervisor

Professor Tobias Hermann

Foundation models are transforming healthcare AI, but their evaluation in high-stakes clinical settings remains inadequate. This project will be developing standardised evaluation protocols for foundation models in intensive care units (ICUs). You will integrate multiple large-scale ICU datasets (MIMIC-IV, eICU, and others) to create a comprehensive benchmark for evaluating foundation models on realistic clinical tasks. This involves harmonising multimodal data from different sources, handling the temporal alignment of asynchronous streams (such as vital signs, laboratory results, and clinical notes), and preprocessing longitudinal patient trajectories. You will then fine-tune foundation models and establish baseline performance on critical ICU tasks, including risk stratification, clinical deterioration prediction, and resource allocation. Your contributions will inform the development of reproducible evaluation standards. The benchmark dataset and code will be released openly to the research community.

Project outcome

You will:

gain a solid background in multimodal healthcare data integration, including handling electronic health records, time-series vital signs, and clinical notes;
become familiar with foundation models (large language models and multimodal models) and fine-tuning approaches for healthcare applications;
develop expertise in benchmark design principles, including dataset construction, evaluation metrics, and reproducibility standards;
implement data preprocessing pipelines and baseline models using modern deep learning frameworks; and
contribute to open-source tools and documentation that will be used by the wider research community.

Entry requirements

You should have, or be studying, a degree in Engineering, Mathematics, Statistics, Computer Science, or a related quantitative discipline. Familiarity with Python is strongly required, with experience in PyTorch, TensorFlow, or similar deep learning frameworks highly desirable. Prior coursework or projects involving machine learning, statistical modelling, or data analysis would be advantageous.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Engineering 09
Cracking the ice: Coupling glacier flow with crevasse formation at continental scale

Primary supervisor

Dr Tim Hageman

Project description

Glaciers don’t just slowly flow downhill and melt, they can break. Predicting when and where crevasses form is key for sea-level rise forecasts and glacier-related hazards. This project builds a next-generation model that couples glacier flow with fracture mechanics, scaling from valley glaciers and small ice-shelves to whole ice sheets. The aim of this specific part of the research is a working coupled scheme that captures both smooth viscous flow and brittle failure.

This project offers several avenues for research projects, depending on your interest/background, including:

computational modelling, extending an existing ice-flow code to capture damage/fractures and experimenting with the numerical schemes used for this; and
geo-data processing, assembling continent-scale datasets and modifying existing models to interface with this data, allowing the existing schemes for small benchmark cases to be applied to real glaciers.

You’ll learn how to integrate fracture physics into large models, test the sensitivity of crevasse development to stress, temperature and geometry, and package results such that they are replicable and re-usable by others.

Project outcome

You will gain hands-on experience in numerical glacier modelling and fracture mechanics, practical coding skills, version control with Git, and reproducible research workflows. You will build and test a prototype coupled flow-fracture module, or develop a continent-scale geo-data pipeline. You will also have the ability to participate in research group meetings, and present the outcomes of your research during them.

Entry requirements

You should have, or be studying, a relevant degree, such as Engineering, Earth Science, Environmental Science, Physics, Mathematics, or Computer Science.

You should have an interest in numerical methods and/or geospatial data and you should be comfortable coding in Python or have a willingness to learn new tools. Experience with numerical modelling, version control (git), and working with geodata would be advantageous.

Engineering 10
Reasoning enhancement in large language models for healthcare applications

Primary supervisor

Professor Tingting Zhu

Project description

Recent advances in large language models (LLMs) have enabled remarkable performance in text understanding, reasoning, and generation. However, their potential for healthcare applications remains limited by challenges such as unreliable reasoning, lack of domain adaptation, and interpretability. This project aims to explore how reasoning enhancement techniques, such as chain-of-thought (CoT) prompting, retrieval-augmented generation (RAG), and fine-tuning with medical datasets, can improve the robustness and transparency of model outputs in healthcare contexts.

You will experiment with open-source models (eg Llama, Mistral) and medical benchmark datasets to evaluate how these methods affect factual accuracy and reasoning quality. The ultimate goal is to contribute towards developing safer, more interpretable AI systems that can support clinicians and medical researchers in decision-making.

Project outcome

You will:

gain a strong background in natural language processing and large language model architectures;
learn how to perform data pre-processing, prompt design, and fine-tuning using medical or biomedical datasets;
develop experience in experimental design, model evaluation, and reasoning analysis;
acquire skills in using tools such as Python, PyTorch, and Hugging Face libraries; and
write a final report comparing your results against baseline methods, with the potential to contribute to a research publication or workshop paper.

Entry requirements

You should have, or be studying, a degree in Computer Science, Data Science, Engineering, Mathematics, or a related quantitative discipline. Familiarity with Python is strongly encouraged, and experience with machine learning or deep learning frameworks (eg PyTorch or TensorFlow) would be advantageous. An interest in artificial intelligence, reasoning, or healthcare applications is desirable.
Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Engineering 11
Digital equity in disaster scenarios

Primary supervisor

Professor Noa Zilberman

Project description

The world is threatened by disasters ranging from severe weather events to pandemics. While technologies like remote healthcare and online education offer critical support, not everyone has equal access to them. This digital divide creates significant inequality during crises, as many cannot use these tools even when available.

As part of this project, you will develop hardware or software mechanisms to improve digital equity for UK residents during disasters. The goal is to enhance accessibility to key digital services. Solutions might include autonomous remote healthcare monitoring, alerting systems, or low-cost, low-latency AI services for the home.

This project is interdisciplinary and covers topics in computer architecture, computer networks, software engineering, and digital equity. AI is an optional, relevant field of study for this work.

Project outcome

As part of this project you will choose to focus on hardware, software or hardware/software co-design. You will explore advanced computing technologies, efficient coding practices, digital equity and potentially AI. You will gain experience in advanced data processing, scripting, power/performance-aware programming, software/hardware development, publishing code and artefacts, and you will help improve digital equity in the UK and our preparedness to future crises. If any publication opportunities will arise from the project, you may be included as a named co-author on that paper.

Entry requirements

You should have, or be studying, a degree in Computer Engineering, Electrical Engineering, or Computer Science. Excellent programming skills in Python or C/C++ or Verilog and basic knowledge in computer architecture or computer networks are required.

Engineering 12
Artificial intelligence for cancer detection in medical imaging

Primary supervisor

Professor Jens Rittscher

Project description

This project will investigate how artificial intelligence can be used to support cancer detection in medical images, for example endoscopic images and videos and other publicly available imaging datasets. The focus will be on designing and evaluating deep learning methods for tasks such as lesion segmentation and detection in realistic data.

You will work with de-identified and public datasets, curate suitable training and evaluation subsets, and implement model training and evaluation pipelines in Python and PyTorch. Supervision will be provided through regular meetings, discussion of experimental plans, and feedback on results, but the work will not follow a step-by-step tutorial format. Instead, you will be encouraged to take intellectual ownership of a well-defined research question and to explore model design choices and performance trade-offs in a systematic manner.

Project outcome

You will gain experience working with medical imaging data (for example, endoscopy images and videos) and additional public datasets, including basic pre-processing and organisation. You will have the opportunity to develop practical skills in data curation, including constructing, cleaning, and documenting training, validation, and test sets. You will implement and train deep learning models in Python and PyTorch for segmentation and/or detection, and evaluate them using appropriate quantitative metrics. You will learn how to design and run controlled experiments (eg comparing architectures, hyperparameters, or data splits) and interpret the results in a clinically relevant context and you will develop scientific communication skills through written documentation and an end-of-project presentation to the research group.

Entry requirements

You should have prior experience in Python and implementing and training deep learning models using PyTorch. You should also have familiarity with basic machine learning concepts, such as, training, validation, test splits, loss functions, optimisation, overfitting, and/or evaluation metrics). Experience with image data and computer vision would be advantageous. No prior medical or biological training is required.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Internet Institute 02
Public interest technology research on AI social biases

Primary supervisor

Professor Luc Rocher

Project description

AI systems often appear neutral but still reproduce prejudice and biases tied to gender, race, age, disability, or socioeconomic status. These effects can emerge in language models, predictive classifiers, recommender systems, privacy techniques designed to safeguard data, or surveillance frameworks used to monitor behaviour. They can lead to real and significant harms by shaping how people are represented and treated.

This project broadly investigates how such technologies encode and amplify social biases and how these dynamics feed back into real peoples’ experiences. The scope is deliberately open: you can examine representational patterns in LLM outputs, measure performance drops in classification systems, study how privacy technologies alter the visibility of minority groups in data, or explore the impact of mass surveillance technologies.

Depending on your interests, you might design a new evaluation method, help us conduct a study with human participants to measure real-world effects, or analyse large existing datasets to identify systematic disparities. You could also work on building an interactive web tool to convey the risks/benefits of using Machine Learning systems to the general public.

Project outcome

You will gain experience analysing the social impact of ML-based systems. You will learn to construct targeted evaluations, interpret model and system behaviour, and/or design empirical studies involving human participants. You will strengthen your programming and data analysis skills.

Entry requirements

You should have, or be studying, a degree in Computer Science, Artificial Intelligence, or a related STEM discipline. A solid understanding of machine learning concepts is essential, along with proficiency in programming languages such as Python. Experience with data analysis and visualisation tools will also be advantageous. Successful candidates will be working on real research projects and so must be self-motivated and fit into a team of researchers from various backgrounds. Good communication skills are important, as you will be expected to present findings clearly to the research team.

Materials 01
High-energy-density Na-rich disordered rocksalt cathodes

Project description

The classical SIR (Susceptible-Infectious-Recovered) model of disease spread assumes a well-mixed population where each individual is equally likely to interact with any other. In reality, social contacts are structured and limited, and during outbreaks, individuals often reduce their number of contacts through, for example, social distancing. This project investigates how contact pruning (the permanent or temporary removal of edges from a contact network) affects disease spread.

We will simulate SIR dynamics on networks with different structures and levels of edge pruning to determine how much contact reduction is necessary to mitigate an outbreak, and which strategies are most effective. The project blends theoretical analysis with simulation, and aims to provide insight into the effectiveness of real-world interventions that correspond to contact pruning.

Project outcome

Through this project, you will gain skills in network modelling and epidemic simulation using Python. You will learn to construct synthetic networks such as Erdős–Rényi, Watts–Strogatz, and scale-free graphs with the NetworkX Python library, and to implement discrete-time SIR models with fixed transmission and recovery rates.

You will also gain experience in quantitative analysis by running simulations across varying network structures and pruning intensities, measuring outbreak size and timing, and comparing outcomes to unmodified networks. In addition, you will learn basic results from percolation theory and mean-field SIR dynamics, and use them to predict critical thresholds for outbreak spread under pruning, and degree distribution and clustering metrics will be compared with observed simulation outcomes.

Entry requirements

You should have, or be studying, a degree in Mathematics, with experience in probability and differential equations. Familiarity with basic concepts in graph theory and dynamical systems is desirable but not essential. Prior experience with Python programming would also be ideal, particularly for data analysis and numerical simulation.

Physics 01
Ultracold atom laboratory

Primary supervisor

Professor Robert Smith

Project description

The project will be based in our ultracold atom laboratory in which we cool erbium and potassium atoms down to nano-Kelvin temperatures to study many-body quantum phenomena such as the recently realised supersolid state. The details of the project you will be working on will be finalised later but could involve design and construction of optical setups for trapping or imaging ultracold atoms, generation of custom magnetic fields for manipulation of atomic properties or numerical simulation of ultracold atom clouds.

Project outcome

You will gain experience in experimental techniques in areas such as optics, electronics, and data analysis, and gain insight into quantum gases. You will write up your work in a short report and if particularly successful you could be a co-author on a publication.

Entry requirements

You should have, or be studying, a degree in Physics (or equivalent).

Physics 02
Accelerating particle accelerators: using reduced-precision floating-point numbers to efficiently model ultra-relativistic beam dynamics

Primary supervisor

Professor Adrian Oeftiger

Project description

Hamiltonian dynamics has broad applicability, from spacecraft trajectories to solar plasmas. Exploiting the geometric properties of Hamiltonian systems extends the horizon for numerical errors in simulations with long time scales, critical for accurate tracking of gravitational systems or proton beams in colliders like LHC. Ultra-relativistic beams in future colliders and existing electron-based X-ray facilities introduce a challenge: synchrotron radiation acts as a random dissipative force, breaking the Hamiltonian framework. Does this make simulations easier or harder to run reliably?

This project investigates how numerical errors accumulate in ultra-relativistic beam dynamics, investigating how these are modified by nonlinear perturbations and chaotic regimes. Conventional methods demand high-precision computing, which is memory-intensive. By studying robustness to low-precision floating-point arithmetic, this work explores efficient use of parallel computation on GPUs. Running many cheap simulations could allow averaging to mitigate numerical noise, potentially offering a practical path forward for future ultra-relativistic beam tracking.

Project outcome

Participating in this project teaches you about the consequences of discretising physics models in order to perform simulations. Such numerical experiments allow for more freedom to investigate physical phenomena. Completing this project successfully will require intuiting the former and developing the latter. While this project focuses on particle accelerators, the gained experience is applicable to any research in computational physics. Beyond developing computational modelling skills, depending on progress in the project, you may have the opportunity to have your results written-up for publication in either an international conference or potentially a scientific journal.

Entry requirements

You should have, or be studying, a degree in Physics, Mathematics, or Computer Science. You must have an interest in computational modelling. A familiarity with coding is useful and knowledge of handling computations on GPUs would be advantageous.

Physics 03
Observing the atmosphere from space

Primary supervisor

Dr Anu Dudhia

Project description

The project involves working within a group specialising in using satellite measurements of the Earth's infrared emission spectrum to retrieve concentrations of a number of different atmospheric gases normally only present in small concentrations (SO2, NH3, C2H6, etc).

The project will involve writing Python code to analyse our data and compare these with other datasets, culminating in a written report and a final presentation to the group. You will be based in an office with other summer project students, with whom they are expected to collaborate, and also participate in weekly group meetings.

Project outcome

You will gain experience in Python coding to access, manipulate and display large datasets, collaborative working within a research group, giving formal presentations, writing a research report.

You will produce a report summarising an investigation into the retrieval of a particular molecule of interest (to be decided nearer the time).

Entry requirements

You should have, or be studying, a degree in Physics or a related STEM subject and have Python programming experience.

Physics 04
Observing volcanic clouds from space

Primary supervisor

Dr Anu Dudhia

Project description

In the Earth Observation Data Group, we have several tools which can be used to detect and quantify information about volcanic clouds (including sulfur dioxide, ash, sulfuric acid) using high spectral resolution measurements. This is critical for understanding the impacts of eruptions on aviation, the atmosphere and climate. We are interested in expanding our tools to incorporate newly launched satellite instruments such as the InfraRed Sounder which will make measurements over Europe every 30 minutes.

You will have the opportunity to work with some of these new instruments – helping to create tools which are used in the future for studying eruptions. Work done during this project will feed into an ongoing collaboration with the UK Met Office where we hope to run these techniques in the long term.

Project outcome

You will gain experience in Python to access, manipulate and display large datasets. You will work collaboratively within a research group and give a presentation at the end of the project. You will also produce a report summarising the results of the study.

Entry requirements

You should have, be studying, a degree in Physics, Earth Sciences or a related STEM subject, and have some Python programming experience.

Physics 05
Hunting for supersymmetric dark matter at the Large Hadron Collider

Primary supervisor

Professor Alan Barr

Project description

Dark matter is one of the most prominent puzzles in modern particle physics. The mystery may be solved by Supersymmetry, a theory that introduces a suite of new fundamental particles which physicists are searching for at the Large Hadron Collider (LHC), the world’s most powerful particle accelerator.

In this project we will explore the hints of supersymmetry that could be present in LHC data. There are several directions the project could take depending on the interests of the student, including applying machine learning methods, statistical techniques and particle physics phenomenology.

Project outcome

Depending on your interests, you will gain experience using machine learning methods, statistical techniques, and/or particle physics phenomenology. The results of the project will motivate future Large Hadron Collider analyses. At the end of the project you will present your findings to our research groups and produce a written report.

Entry requirements

You should have, or be studying, a degree in Physics. Experience with Python, or a desire to learn, is essential. Knowledge/skills in statistics and C++ may be advantageous but are not essential.

Statistics 01
Machine Learning and AI for SARS-CoV-2 Mpro Inhibitor Discovery

Primary supervisor

Professor Garrett Morris

Project description

You will learn how to apply the latest machine learning and AI technologies to help discover new inhibitors of a key drug target in SARS-CoV-2, the virus that causes COVID 19. By training models on binding data and 3D atomic structures of inhibitors of SARS CoV-2 main protease, you will advance our understanding of how to block viral maturation and how to develop new drugs to treat COVID-19.

Project outcome

You will explore data from the COVID Moonshot project to develop a variety of classical ML models and more advanced methods such as Graph Neural Networks, Atomic Environment Vector-based models, and molecular transformers.

Entry requirements

You should have, or be studying, a degree in Computer Science, Statistics, Engineering, Bioinformatics, or another relevant discipline. This project is suitable for those seeking to improve their knowledge of practical machine learning and complex deep learning architectures, and learn the application of such models to protein-small molecule complex structures. You should have experience of machine learning and programming with Python. Experience with the deep learning framework PyTorch would be beneficial.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Statistics 02
Understanding the Effects of Data Quantity and Label Noise on Machine Learning Models in Drug Discovery

Primary supervisor

Dr Fergus Imrie

Project description

This project explores how data quality and quantity affect the performance of machine learning models in drug discovery. In particular, it focuses on QSAR modelling, where computers learn to predict how strongly different molecules might interact with a biological target. Experimental data is often limited and can contain errors (known as label noise), which can reduce model reliability.

Using publicly available data, the project will systematically vary the amount of training data and level of label noise to observe the impact on the performance of machine learning models. The project could then investigate the use of data-centric methods (approaches that focus on the data and its quality, or lack thereof, rather than changing the model) to reduce the impact of noisy measurements. The ultimate aim is to develop practical guidance and techniques for building robust QSAR models when working with imperfect experimental data.

Project outcome

You will gain hands-on experience in computational data analysis, machine learning, and cheminformatics tools, and use them to build and evaluate QSAR models. This project will also develop your ability to visualise results, interpret model behaviour, and communicate scientific findings. Throughout the project, you will be supported in understanding key concepts in machine learning for drug discovery.

You will present your findings in an internal meeting at the end of the project. You may have the opportunity to contribute to a publishable research project (either standalone or as part of other projects), for which you would be included as a named co-author.

Entry requirements

You should have, or be studying, a degree in Computer Science, Statistics, Engineering, Biochemistry, Bioinformatics, or another relevant field.

This project is suitable for those seeking to improve their practical knowledge of machine learning, and learn to apply such models to real-world use cases, such as drug discovery.

You should have experience of machine learning and programming with Python. Experience with machine learning frameworks, such as scikit-learn or PyTorch, or molecular data would be beneficial, but is not required.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Statistics 03
Mathematical modelling of the drivers and dynamics of dengue epidemics

Primary supervisor

Dr Cathal Mills

Project description

Dengue is a mosquito-borne disease, whose distribution and incidence are expanding and increasing globally. With environmental sensitivities across the mosquito and virus, dengue primarily affects low- and middle-income countries in tropical and sub-tropical regions. However, climate change, urbanisation, and globalisation coalesce to impose growing public health burdens on resource-limited settings. Faced with this complex public health and environmental challenge, decision-makers need timely and reliable information to respond appropriately.

In this project, you will aim to improve our ability to understand and control dengue outbreaks. Recent mathematical models offer avenues to investigate the time- and space-varying causal mechanisms of large dengue outbreaks. We can now describe dengue virus transmission richly using partial differential equations across populations of infected and infectious mosquitoes and humans. However, it is not yet understood how best to implement such frameworks in real time, how different environmental factors interact with each other and with human and biological confounders, or how to incorporate the effects of stochasticity.

Using numerical and analytical approaches, the aim of this project will be to use mathematical modelling to contribute meaningfully to understanding and controlling dengue epidemics.

Project outcome

You will have the opportunity to develop numerical and analytical approaches for investigating systems of partial differential equations. These approaches will blend mathematical biology with modern machine learning and statistical inference techniques. You will have the opportunity to present your results to the group throughout the internship. At the end of the internship, you may have the opportunity to contribute to important ongoing or new publishable work, for which co-authorship would be recognised.

Entry requirements

You should have, or be studying, a degree in Mathematics or Statistics and be familiar with a programming language of your choice.

Statistics 04
Anytime-valid inference via coin betting

Primary supervisor

Professor Patrick Rebeschini

Project description

This project investigates anytime-valid statistical inference through the lens of the coin betting framework. Traditional statistical tests assume a fixed sample size, but modern data analysis often involves continuous monitoring and adaptive decision-making. Coin betting offers a principled approach to address this challenge by constructing supermartingales—wealth processes that remain valid under the null hypothesis—leading to e-values and confidence sequences that maintain correct coverage at all times.

You will explore how coin betting can be used to design new sequential testing and estimation procedures with strong finite-sample guarantees. Potential directions include developing e-processes for high-dimensional or dependent data, analysing trade-offs between adaptivity, power, and confidence width, and studying the duality between statistical evidence and wealth accumulation. The project emphasizes rigorous theoretical analysis, with optional simulations to demonstrate the robustness and practicality of the proposed inference methods.

Project outcome

You will gain experience in modern statistical methodology, with a focus on sequential inference and the coin betting framework. You will be trained in how to construct and analyse supermartingales and e-values, derive confidence sequences, and apply these tools to real-world settings such as A/B testing used eg for website/game design optimisation. You will also develop skills in theoretical analysis and mathematical proof, supported by computational experimentation using Python to illustrate key results.

Throughout the project, you will receive guidance on communicating complex theoretical ideas clearly and rigorously. At the end of the project, you will present your findings in an internal seminar and prepare a short, written report summarizing your contributions. If your theoretical or methodological developments contribute to future research outputs, you may be acknowledged or included as a named co-author on a resulting publication.

Entry requirements

You should have, or be studying, a degree with elements of probability theory, ideally at the level of martingales. Familiarity with Python would be advantageous.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Life and Medical Sciences

Entry requirements

You should have, or be studying, a degree related to Biological Sciences, Molecular Biology, Cell Biology, or Biochemistry with a strong interest in plants. Having some lab experience and the ability to work independently will be useful, although any required training and support will be provided.

Biology 04
From Wuhan to the world: A genomic reconstruction of early SARS-CoV-2 spread

Primary supervisor

Dr Mahan Ghafari

Project description

This project will explore how SARS-CoV-2, the virus causing COVID-19, spread globally in the early stages of the pandemic using publicly available viral genome sequences. You will learn how to assess sequence quality, align genomes, and place them on a phylogenetic tree to see how samples cluster by time and place. By examining these clusters, we will infer patterns of introduction, early transmission, and how new mutations emerged and rose in frequency. The project will use computational tools common in genomic epidemiology (eg sequence QC, multiple sequence alignment, tree visualisation) and will culminate in a short report on what the data reveal about early viral evolution and spread.

Project outcome

You will learn how to work with real SARS-CoV-2 genomic data, including downloading public sequences, performing basic quality control, conducting evolutionary analysis, and exploring how sequences group on a phylogenetic tree. You will gain hands-on experience with core tools in genomic epidemiology (sequence alignment, tree visualisation, and simple mutation tracking) and learn how these analyses inform our understanding of viral evolution and global spread. You will also practice summarising genomic findings for a non-specialist audience and present your results to the group at the end of the internship.

Entry requirements

You should have, or be studying, a degree in a relevant subject such as Biology, Biomedical Sciences, Biochemistry, Genetics, or a quantitative discipline (eg Mathematics, Statistics, or Computer Science). Basic computer skills and an interest in working with data are encouraged, but prior experience in bioinformatics is not mandatory as training will be provided. A keen interest in viruses, infectious disease, or how SARS-CoV-2 evolves and spreads is essential. The project will suit a motivated student who is curious, comfortable learning new software, and able to work carefully with real-world data.

Biology 05
Seeing nature through citizen eyes: analysing flower colour and spread in an invasive species

Primary supervisor

Professor Rob Salguero-Gomez

Project description

This project will explore patterns of flower colour and distribution in an invasive plant species using photographs submitted by the public through citizen science platforms. You will collect, organise, and analyse image data to investigate how flower colour varies across regions and whether environmental factors influence these patterns. You will gain hands-on experience in data management, programming, and the use of accessible artificial intelligence tools to help automate image labelling and analysis. The project also involves spatial data exploration to understand distribution trends. Through this work, you will develop transferable coding and research skills, deepen your understanding of invasive species ecology, and appreciate the value of citizen science in advancing environmental research and engaging the public in scientific discovery.

Project outcome

You will gain experience in handling and analysing large citizen science image datasets, learning how to extract meaningful information about flower colour and distribution patterns. You will be trained in programming (using Python or R), data cleaning, image classification, and introductory machine learning methods to support automated labelling. You will also learn principles of programming reproducibility and how to create and maintain your own coding portfolio using GitHub.

By the end of the project, you will have produced a short written report and co-created a video explaining your findings. You will also present your results to the research group and may be acknowledged in future research outputs arising from the project.

Entry requirements

You should have, or be studying, a degree in Biology, Environmental Science, Geography, or a related quantitative subject such as Data Science or Computer Science. Some experience with data handling, programming (preferably in Python or R), and an interest in applying computational tools to ecological questions would be advantageous. Familiarity with image analysis, machine learning, or GIS software is desirable but not essential, as training will be provided. You should be curious, motivated to learn new analytical techniques, and comfortable working independently with guidance. An enthusiasm for biodiversity, citizen science, and communicating research to the public will help you get the most from this project.

Biology 06
How does cacao and coffee farming impact soil organic carbon?

Primary supervisor

Dr Joseph Poore

Project description

Changes in soil organic carbon (SOC) stocks on agricultural land can significantly influence CO2 emissions or removals. Adopting land management practices that increase SOC offers opportunities to reduce atmospheric CO2 and offset the carbon footprint of agricultural production. To better understand these effects, we are compiling the world’s largest harmonised archive of SOC data, linked to production practices and land-use history, enabling identification of drivers of SOC changes.

This project focuses on cacao and coffee systems globally, covering 11.7 and 12.2 million hectares, respectively, in 2023. Both crops are major drivers of deforestation in tropical regions, including Indonesia, Côte d’Ivoire, and Brazil. The internship will investigate SOC dynamics when forests are converted to plantations and assess how agroforestry can mitigate SOC losses compared with monocropping.

The main task is uploading SOC sampling data to the HESTIA platform, evaluating the effects of land management and land-use changes, with automatic calculations of associated CO2 emissions or removals and their broader life-cycle impacts.

Project outcome

You will gain knowledge of the role of soil organic carbon sequestration in climate change mitigation efforts in the agricultural sector and how greenhouse gas emissions linked to agricultural activities are quantified and modelled using Life Cycle Assessment methods. You will be familiarised with cacao and coffee production systems and their broader environmental impacts. You will gain experience in conducting literature reviews, data mining and data analysis. You will also gain skills in Excel, Git and JSON schemas.

Entry requirements

You should have, or be studying, a degree in a quantitative discipline, such as Biology, Chemistry, Engineering, Environmental Science, Mathematics, Physics.

Biology 07
Signals and strategies: how fish communicate to manage cooperation and conflict

Primary supervisor

This project aims to run biodiversity surveys across schools in Oxford to provide a baseline assessment, which can yield meaningful ecological data and be used as a benchmark for future outreach and public engagement projects. Information gained from this pilot will be shared with participating schools, as well as the Healthy Ecosystem Restoration in Oxfordshire network and the British Ecological Society.

Project outcome

By the end of the internship you will have:

gained experience in conducting biodiversity surveys across a range of taxonomic groups;
contributed to survey planning and setup; collected data contributing to a potential future publication, which you would be asked to co-author;
gained communication skills by presenting your findings to colleagues, teachers and students; and
gained project management skills and experience of working as part of a research team.

Entry requirements

You should have, or be studying, a Biology-related degree, and have an interest in ecology and conservation work. Some species identification skills would be very useful, although help will be provided in expanding these. You will be a self-starter with the ability to work independently.

You will be required to undergo a DBS check, since you will be working on school grounds.

Clinical Medicine 01
International health workforce recruitment in the UK

Primary supervisor

Dr Yingxi Zhao

Project description

The UK’s health and care services face major challenges in recruiting and retaining sufficient staff. Currently, one in five NHS or social care workers in England is internationally recruited, a figure that is expected to rise. Private recruitment agencies are playing an increasingly central role in facilitating the international migration of health workers, particularly nurses, against the backdrop of a global workforce shortage. This internship, building on work by the Health Systems Collaborative team on UK and global health workforce issues, will involve a series of literature, policy reviews, and web-based content analysis to map recent policy changes, identify key actors, and update the evidence base on the experiences of internationally recruited staff.

Project outcome

In this project, we will perform in-vitro experiments to investigate how different bacterial strains behave when attached to microplastics. Using a diverse selection of materials from our in-house microplastics library, we will create a mock bacterial community for inoculation. You will perform growth experiments combining this community with various microplastics and assess resistance related parameters using classical, molecular, and computational microbiology approaches.

Project outcome

During the project you will work with microbiologists and computer scientists, developing skills in microbiology, assay development and data analysis. The aim of the project will be to produce a summary scientific report on the findings of your experimental evaluation and present this in one of our research laboratory meetings. Depending on the work undertaken and findings this could also be submitted as a conference abstract and/or contribute to a scientific publication.

Entry requirements

You should have, or be studying, a degree in Microbiology or a related discipline. Previous laboratory experience in culture, DNA extraction and sequencing is desirable.

Clinical Medicine 05
Generating second-generation human lymphoid organoids as a model to study adaptive immune responses

Primary supervisor

Dr Delaney Dominey-Foy

Project description

This project will explore how human immune cells interact and develop within complex three-dimensional tissue models. Using advanced lymphoid organoids, which are miniaturised systems that mimic the structure and function of human lymphoid tissue, you will investigate how T and B cells coordinate adaptive immune responses. You will gain practical experience in human cell culture, microscopy, and data analysis, contributing to ongoing efforts to establish next-generation organoid models for studying human immunity.

Project outcome

You will gain hands-on experience in human tissue culture, including the generation and maintenance of advanced lymphoid organoids. You will be trained to use confocal microscopy to image immune cell interactions and to analyse these images using AI- and machine learning–based tools. You will also perform and interpret multicolour flow cytometry to identify and quantify immune cell populations.

Alongside these technical skills, you will develop your ability to handle, visualise, and statistically analyse complex biological datasets. You will also gain experience in presenting your findings, both in group meetings and in a final research presentation.

By the end of the project, you will have a deeper understanding of human immunology and how adaptive immune responses can be studied using next-generation organoid models. If your data contributes to a future publication, you may be acknowledged or included as a co-author.

Entry requirements

You should have, or be studying, a degree in a relevant Biological or Biomedical Science, including courses that contain content on immunology, cell biology, biochemistry, or a related field. An interest in human immunology and experimental biology is essential.

Previous laboratory experience is helpful but not required as full training will be provided. Confidence in handling data and a willingness to learn computational approaches (for example, image or flow cytometry analysis) would be advantageous.

You should have good organisational skills, attention to detail, and enjoy working collaboratively as part of a research team. Curiosity, motivation to learn new techniques, and an enthusiasm for experimental discovery are the most important qualities.

Clinical Neurosciences 01
Developing shark antibodies as biomarkers for peripheral nerve injury

Primary supervisor

Dr Alexander Davies

Project description

Neuropathic pain is a chronic condition that occurs because of injury or disease of the nervous system. It is thought to affect up to 10% of the adult population in the UK, yet diagnosis can be challenging. Identifying nerve damage in individuals with chronic pain would help tailor treatments and enhance our mechanistic understanding of the condition.

This project aims to develop small peptide molecules – shark antibodies - as biological tools to recognise injured sensory neurons in mice and humans. We will validate them as potential novel biomarkers for nerve injury.

You will be trained in wet-lab techniques relevant to the project at the time, which may include immunohistochemistry, enzyme-linked immunosorbent assay (ELISA), live cell-based assay and confocal microscopy.

In the longer term this work could help develop therapeutic interventions that target injured nerves specifically and avoid the side-effects of current medications for neuropathic pain.

Project outcome

You will be part of a small research team working in collaboration with an industrial partner employing proprietary technology to address this biological and therapeutic challenge. As well as being taught laboratory skills relevant to biological research more generally, you will learn how to design, execute and analyse research experiments. You will attend weekly lab and journal club meetings, as well as monthly project meetings with our industrial partner. At the end of your project you will have the opportunity to present your findings at a large group meeting.

Entry requirements

You should have, or be studying, a Biology or Biomedical Science-related degree with and have an interest in nervous system function in the context of health and disease. You will be able to input data into a spreadsheet and perform basic statistical analysis. You will appreciate the core principals of experimental design and scientific report writing. You will understand the basic principles of neural anatomy, antibody chemistry, and fluorescence microscopy. Experience in a biological laboratory is useful but not essential.

Clinical Neurosciences 02
Enhancing cognitive impairment detection and inclusivity in stroke care

Primary supervisor

Dr Sam Webb

Project description

Aim: There is no one assessment that can accurately flag impairments in thinking skills following a stroke that:

is suitable for any time post-stroke;
is suitable for different severity of strokes (mild to severe);
can account for different impairments in vision, attention; motor abilities, and communication; and
can be completed in a brief as time as possible.

We designed such a test in an app that does all of these things, and we are currently assessing its accuracy and validity in multiple stroke settings.

Method: Administering neuropsychological assessments to stroke survivors who live in the community (in and around Oxfordshire), scoring assessments, entering data, and brief data analyses. Training in test administration and data curation/analysis will be provided.
Project outcome

You will be trained in neuropsychological assessment methods and scoring, as well as in data analyses using R. You will have the opportunity to contribute to lab meetings and gatherings during your internship. You will also gain experience in a translational neuropsychology research group, learning about the different types of research conducted and future pathways.

Entry requirements

You should have, or be studying, a degree in Psychology or closely related field, such as Neuropsychology. This project would suit someone who is interested in studying and/or progressing into clinical (neuro)psychology.

Clinical Neurosciences 03
Modelling neurodevelopment with cerebellar organoids

Primary supervisor

Professor Esther Becker

Project description

This project aims to understand how the brain’s immune cells, called microglia, help shape the development of the cerebellum, a key brain area that is important for motor control but also cognitive and emotional processes. We are using human induced pluripotent stem cells (iPSCs) to generate cerebellar organoids and will investigate how microglia and cerebellar neurons interact in this novel model system. Interesting questions that could be explored in the project include:

Do microglia influence the development and maturation of specific neuronal subtypes in the cerebellum?
How does the cerebellar microenvironment shape the identity and function of co-cultured microglia?

Lab techniques to address these questions will include tissue culture, molecular techniques such as qRT-PCR, and immunostaining.

There might also be the possibility to undertake functional analyses using multi-electrode arrays.

Project outcome

You will gain experience in experimental design, tissue culture, molecular and imaging techniques, and data analysis. You will have regular meetings with your supervisors to discuss your project and plan next steps. At the end of the project, you will present your findings back to the group at our lab meeting.

Entry requirements

You should have, or be studying, a degree in Biomedical Sciences or equivalent, such as Biology, Biochemistry, or Pharmacology. An interest in neuroscience would be advantageous. Training in all required techniques will be provided, but knowledge of experimental techniques and wet-lab experience would be useful.

Clinical Neurosciences 04
Immunohistochemical characterization of LGI1-AAV in a neuropathic pain model.

Primary supervisor

Professor John Dawes

Project description

Neuropathic pain arises from damage to the nervous system and is associated with overactive neurons in the dorsal root ganglia (DRG) and spinal cord. Current treatments are often ineffective, highlighting the need for new approaches. Leucine-rich glioma-inactivated 1 (LGI1) is a secreted protein that regulates neuronal excitability (Fels et al., 2021). Autoantibodies against LGI1 have been linked to pain syndromes, and reducing these antibodies improves symptoms (Gadoth et al., 2017; Ramanathan et al., 2021). Using adeno-associated virus (AAV9) gene delivery, we increased LGI1 expression in a mouse model of nerve injury and found reduced pain sensitivity (Farah et al., 2024).

To understand how LGI1 works, we will use immunohistochemistry and confocal microscopy to study its expression and viral transduction in specific neuronal populations from spinal cord and dorsal root ganglia tissue. This project aims to identify LGI1’s mechanisms of action and evaluate its potential as a novel gene therapy for neuropathic pain.

Project outcome

You will gain hands-on experience in neuroscience and molecular biology techniques, focusing on neuropathic pain mechanisms. You will be trained to handle and analyse experimental data from immunohistochemistry, learning to identify specific neuronal populations using confocal microscopy. You will also learn the principles of viral vector design and gene delivery, as well as how to interpret and present quantitative results using appropriate statistical methods. Additionally, you will gain insight into how preclinical models are used to investigate therapeutic strategies for nervous system disorders. The intern will present their findings in a lab meeting at the end of the placement and, if their work contributes to future publications, may be acknowledged or included as a co-author.

Entry requirements

You should have, or be studying, a degree in Neuroscience, Biomedical Sciences, Physiology, Pharmacology, or a related discipline. An understanding of basic neurobiology and molecular or cellular biology techniques would be advantageous. Prior experience with laboratory work would be useful but is not essential, as full training will be provided. You should have strong organisational skills, attention to detail, and enthusiasm for experimental research. The ability to work carefully, follow protocols, and communicate findings clearly will be important for success in this project.

Clinical Neurosciences 05
Estimating sleep from wearable devices

Primary supervisor

Professor Simon Kyle

Project description

Wrist-worn actigraphy is commonly used in sleep research and clinical sleep medicine to estimate sleep and rest-activity rhythms. Actigraphy data are typically combined with patient-reported sleep diaries and manually scored by a clinician or researcher. However, the inclusion of actigraphy in large-scale cohort studies (involving 100,000+ participants) has led to the development of automated analysis approaches, enhancing efficiency and reliability. Nevertheless, it is unclear how sleep outcomes might differ between automated and manual scoring approaches.

In this project, we will explore how sleep parameters may differ when data are manually scored, with the aid of a sleep diary, compared to outcomes from automated scoring. You will set-up actigraphy devices, collect data using these devices, and then analyse how the data compares across scoring methods. You will have the opportunity to learn more about sleep research methods, how to score sleep data, and plan and implement your own data analysis plan.

Project outcome

You will have the opportunity to learn about a range of wearable and nearable technologies used within sleep research. The focus will be on analysing data from a wrist-worn actigraphy device, and investigating how sleep parameters differ across manual and automated scoring. You will also have the opportunity to learn about polysomnography, regarded as the ‘gold standard’ for sleep research, and learn how to set-up and score sleep polysomnography data.

You will be expected to explore a wide range of literature around actigraphy and its scoring, collect your own data, create and implement a statistical analysis plan, and write a report of what you find. Additionally, we would like you to present your findings at a lab meeting.

Entry requirements

You should have, or be studying, a degree in Neuroscience, Psychology, or a related discipline. This project will suit someone with an interest in sleep and/or wearable technologies. Experience with data analysis is beneficial, but support and training will be provided if needed.

Clinical Neurosciences 06
Record and analyse the spontaneous neuronal firing in a humanised live neuron recording system

Primary supervisor

Dr Pao-Sheng Chang

Project description

Neuropathic pain is affecting 9% of people worldwide, but many promising pre-clinical models fail to be translated into effective treatments. To better investigate the paroxysmal pain in human, we have developed a bespoke humanised live neuron recording system to measure the level of spontaneous firing in human sensory neuron. The system is applied to understand the causal relationship between co-culture cells or inflammatory mediators with human sensory neuron by measuring the level of spontaneous neuronal firing. We can then relate the cell-based measurements to the paroxysmal pain in clinics for developing better pain treatments.

Project outcome

You will experience live neuron recording conduction and analysis whilst also gaining the practical experience in maintaining human iPSC-derived sensory neuron in culture dish apart from other fundamental wet-lab practices. We aim to build up your experience and confidence in the lab with providing guidance for future study and career trajectory.

You should have, or be studying, a degree in a relevant discipline, such as Biology, Biomedical Sciences, Biochemistry, Pharmacology, or a related discipline and have a basic understanding of immunology, particularly T-cell biology and antigen presentation.

The ability to follow experimental protocols and maintain accurate labs is essential. Laboratory skills such as sterile cell culture techniques, pipetting, and handling biological samples would be advantageous, but is not a requirement. You should be motivated to learn new techniques and engage with interdisciplinary approached combining wet-lab and computational work.

Medicine 03
Investigating T–B cell interactions in vaccine responses using human tonsil organoids

Primary supervisor

Dr Nicholas Provine

Project description

Human tonsil organoids have recently emerged as a promising model for studying vaccine-induced immune responses in vitro. Our group has validated this system for modelling responses to the adenovirus-vectored vaccine ChAdOx1 nCoV-19, better known as the Oxford–AstraZeneca COVID-19 vaccine, and identified key cytokine pathways involved in B and T cell responses. The next step is to determine how well this model reproduces the interaction between T cells and B cells, which is essential for classical antibody production in vivo.

In this project, you will manipulate T cell function in the tonsil organoid system using specific antibodies and assess the effect on B cell activation and antibody secretion. Immune responses will be analysed using flow cytometry and ELISA. This work will advance understanding of the tonsil organoid model and refine its use for studying vaccine-induced immunity.

Project outcome

You will be trained in fundamental immunology techniques, including cell culture using primary human lymphoid cells, flow cytometry for characterising immune cell activation, and ELISA assays for measuring antigen-specific antibody responses. You will also gain hands-on experience in experimental design, data recording, and data interpretation.

Through this work, you will develop practical laboratory skills and a deeper understanding of vaccine immunology. You will be encouraged to present your findings in a group meeting, and if your data contributes to a future publication, you will be acknowledged as a named co-author.

Entry requirements

You should have, or be studying, a degree in a biological or biomedical science subject, such as Immunology, Microbiology, or Biochemistry. An interest in immunology or infectious diseases would be beneficial. You should have good attention to detail, be willing to learn new laboratory techniques, and have a careful and methodical approach to experimental work. Prior laboratory experience is not required, as full training will be provided.

Medicine 04
Decoding the premature gut: using spatial transcriptomics to gain insights into necrotising enterocolitis (dry lab)

Primary supervisor

Dr Agne Antanaviciute

Project description

Necrotising enterocolitis is a severe intestinal disease affecting premature infants, where parts of the gut become inflamed and die. Despite medical advances, the cause remains poorly understood, making it difficult to diagnose and treat. In severe cases, the disease can progress rapidly, leading to holes forming in the intestinal wall that require surgery.

The Antanaviciute and Simmons labs, based at the Weatherall Institute of Molecular Medicine, investigate intestinal development and disease using cutting-edge techniques such as spatial transcriptomics, which reveal how genes are active in different parts of the gut.

This project offers a rare opportunity to analyse high-dimensional spatial data from human neonatal tissue – an invaluable and limited resource. You will learn to analyse and identify cellular changes that drive disease progression in premature infants with Necrotising Enterocolitis, which may contribute to identifying early biomarkers to improve diagnosis and treatment.

Project outcome

You will receive training in computational tools used to analyse high-dimensional spatial transcriptomics data – a cutting-edge approach recognised by Nature Methods as 'Method of the Year' in 2020. Although still in its early stages, this technology holds immense potential for uncovering complex biological processes. You will work within an interdisciplinary team of biologists, clinicians and computational scientists, gaining a holistic understanding of gut development and disease and clinical research pipeline. While this is a dry lab project, there may be opportunities to observe or contribute to molecular biology experiments if desired. If any aspect of analysis is included in future publication, you may be included as a named co-author.

Entry requirements

You should have, or be studying, a degree in Biology, Biochemistry, Biomedical Science, or a related discipline. As it is a fully computational project, some experience in R would be an advantage.

Medicine 05
AI deep learning for clinical research

Primary supervisor

Professor Qiang Zhang

Project description

AI deep learning is transforming the world in many aspects, but its real-world clinical applications have been hindered by the knowledge gap between machine learning scientists and clinicians. We actively address this by developing AI algorithms next to clinical doctors at an interdisciplinary clinical research unit. We offer opportunities to work with a cross-disciplinary team to gain experience in developing deep-learning solutions for unmet clinical needs. This may include data pre-processing, neural network design, data analysis, and method validation.

You will have the chance to observe real-world clinical MR scans at the Division of Cardiovascular Medicine, and access to computing facilitates at Oxford Big Data Institute.

Project outcome

You will gain domain knowledge of both deep learning and cardiovascular imaging, skills in medical data processing, neural network design in Python, and valuable experience in developing AI algorithms in clinical research settings.
Entry requirements

You should have, or be studying, a degree in Computer Science or Engineering. You should have experience in machine learning and coding in Python.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Medicine 06
Developing a simulation framework for clinical hyperpolarized MRI experiments

Primary supervisor

Dr James Grist

Project description

This project will focus on developing a computer simulation that can be used to describe, and more importantly optimise, metabolic imaging experiments in humans. You'll use data already present in the lab, as well as fundamental physics mathematics, to make this framework as true to real life as possible. You'll use Python or Matlab to code the project and be well-supported in your time here.

Project outcome

You will be trained in fundamental MRI physics, python, Matlab, and image processing. You will have the opportunity to engage in lab meetings and experience and understand the work being performed by other members of the lab. You will be involved in research discussions and the day to day life of the lab. You'll be able to present your work at lab meetings and gain experience in scientific presentation.

If completed, you will be able to publish this work.

Entry requirements

You should have, or be studying, a degree in Computer Science, Engineering, or Physics. Experience in Matlap or Python is essential.

Medicine 07
Super-resolution imaging of cell adhesions in gut patient organoids

Primary supervisor

Dr Karina Pombo-Garcia

Project description

Our gut ages, which has implications for the cellular organization of proteins involved in cell-cell adhesion. Cell-cell adhesions are like the building blocks of a house, and the proteins responsible for holding them together act as the glue that allows our organs to function. In this project, you will have the opportunity to visualize, at very high resolution (on the nanometer scale), the distribution and organization of cell adhesion proteins in a gut-like model called gut-organoids. We will use patient-derived organoids from individuals of different ages to map how these proteins assemble as we age.

You will learn and use super-resolution microscopy with fluorescence proteins to visualize them in human tissue. Additionally, you will get familiar with handling the organoids and observing how they phenotypically change over several days in culture. While not required, it would be ideal if you have some basic knowledge of coding, such as Python, to assist in processing the images.

Project outcome

You will gain experience in several key cell biology techniques as well as being immersed in an active research environment. This will enable broadening of expertise, as well as help you derive a better understanding of studying and working in a research laboratory.

The project itself will allow for training in cell culture and aseptic techniques. You will also be trained on key super-resolution imaging microscopy STED, as well as gut organoid development and maintenance.

There will also be the opportunity to work in a lab with a range of experience from junior to senior scientists who are willing to advise and mentor. If successful, the data and resources generated will have a real-world impact on ongoing research with your contribution credited accordingly if work with their direct involvement is published.

Entry requirements

You should have, or be studying, a degree in Biology, Biochemistry, or Medicine. Basic understanding of cell biology and microscopy, alongside familiarity with Python, would be advantageous.

Medicine 08
The extracellular vesicle profile of inherited cardiomyopathy

Primary supervisor

In the Gerard lab, we aim to understand how Tregs control T cells not to attack our self, while allowing them to fight infection. To do so, we study where both Tregs and conventional T cells are found and how they influence each other during infection. We believe that Tregs are segregated from the site of infection, which is important to allow conventional T cells that can recognise infection to get activated and kill infected cells.

In this project, you will shadow and participate to our effort to characterise the location and communication between Tregs and conventional T cells during infection using techniques such as imaging or flow cytometry. You will also learn how to generate Tregs.

Project outcome

You will learn cell culture and imaging and get a deeper understanding of the immune system.

Entry requirements

You should have, or be studying, a degree in Medicine, Biology, or Immunology.

NDORMS 02
Systems biology of pain in chronic human inflammatory disease

Primary supervisor

Professor Mark Coles

Project description

This project will involve spatial transcriptomics of biopsies from patients with immune mediated inflammatory disease and relating the spatial gene expression to pain scores. The techniques include the application of a python package to analyse cellular segmentation and quantify the gene transcription on a single cell basis and then to use this information to quantify the structures in the tissues. The project will involve learning python and other data analytical techniques. The project aims to identify cells within the tissue biopsy, quantify gene expression in individual cells and then use this information to develop a map of cellular organisation and function and how this relates to pain.

Project outcome

You will undertake spatial data analysis of human tissue sections from inflammatory disease using with the aim of generating evidence for molecular pathways that might related to patient measured pain. The analysis will identify key cell types involved in pain and the phenotype of those cell types comparing patients with high pain to controls.

Entry requirements

You should have, or be studying, a degree in Mathematics, Physics, Computer Science, or another relevant discipline, and be interested in applying your skillset to a biological problem.

NDORMS 03
Real-world evidence on the safety and effectiveness of novel medicines in cardiovascular and metabolic diseases: A systematic review

Primary supervisor

Dr Annika Jödicke

Project description

Understanding how well the new medicines for cardiovascular and metabolic disease work, and how safe they are in real-world settings is essential. This project will focus on 'real-world evidence' (RWE), referring to studies using data gathered in routine care settings, such as by general practitioners in primary care, or in specialist settings of hospitals. Systematic reviews of the literature provide a comprehensive overview of the published knowledge at a time. This is particularly important to identify gaps for future.

You will begin by developing a protocol for a literature review, including a search strategy and criteria for including or excluding studies. With guidance from supervisors, you will then review and discuss the identified studies with the supervisor team, and write a summary of the findings that will inform future research. This work will provide valuable experience in research methodology, with the potential to published as part of a project report or scientific manuscript.

Project outcome

You will have the opportunity to gain experience in conducting literature reviews using a systematic search strategy. We will support you in developing a protocol for the review, work with specialists at the Bodleian Libraries to develop a search strategy, teach you how to screen titles and abstract, and extract relevant information from the scientific publications we shortlist.

We aim to write a report on the literature review that will be published alongside the protocol, and, where possible, include the findings from the literature review for the introduction/discussion sections of future research publications or project reports.

Entry requirements

You should have, or be studying, a degree in a medical-related field, eg Public Health, Nursing, Pharmacy, Epidemiology, Medicine, or Statistics. An interest in reading medical literature and are interested in learning how to conduct literature reviews would be advantageous.

NDORMS 04
Bridging clinical trials and real-world data: Replicating and characterising clinical trial population from real-world healthcare databases

Primary supervisor

Dr Martí Català

Project description

Join a dynamic and multidisciplinary team in the Health Data Sciences group at NDORMS, University of Oxford. Our team comprises experts from various fields, including medicine, pharmacy, mathematics, statistics, and computer science, who collaborate to use real-world data (RWD) from hospitals, general practices, and disease registries to improve public health.

While essential for testing new treatments, clinical trials are conducted in controlled settings with carefully selected patient populations. This limits their ability to fully represent how treatments perform in the real world. By contrast, RWD reflects the broader population and real healthcare practices, offering insights that clinical trials might miss, such as unobserved side effects or how treatments are used in routine care. However, since RWD was not originally collected for research purposes, careful processing and analysis are required to ensure reliable scientific evidence.

In this internship, you will reproduce and characterise (such as analysing the survival of patients with a specific disease or comparing the effectiveness of a drug) a population of interest using RWD sources, including data collected from general practices, to answer research questions.

Project outcome

This internship provides comprehensive training in R programming, big data handling, clinical epidemiology, and biostatistics. You will have the opportunity to build your presentation skills by regularly presenting your work in team meetings, with potential for co-authorship on project publications. These skills are highly valued in both academia and industry (particular in pharmaceutical related fields), providing a solid foundation for future job applications and career development.

Entry requirements

You should have, or be studying, a STEM-related degree. Familiarity with R programming is desirable.

NDORMS 05
Uncovering how genetic variation shapes gene activity in single cells

Primary supervisor

Dr Yang Luo

Project description

This project explores how genetic variation influences gene activity at single-cell resolution. Using data from a newly generated single-cell study, you will investigate how genetic differences shape gene regulation across cell types. Possible directions include visualising genetic associations, linking variants to regulatory mechanisms, or applying simple AI or machine-learning approaches to identify patterns in the data.

The project will use computational and bioinformatics tools such as R, Python, and online genomic databases. It offers hands-on experience with real genomic data and introduces how AI-driven analysis can enhance our understanding of gene regulation and its role in human disease.

Project outcome

You will gain hands-on experience working with genomic and single-cell data, using tools such as R, Python, and online databases for data analysis and visualisation. You will develop practical skills in bioinformatics workflows, data integration, and the use of basic AI or machine-learning methods to identify patterns in complex datasets.

By the end of the project, you will understand how genetic variation influences gene activity and feel confident working with large-scale biological data. You will present your findings to the research group, and if your analysis contributes to ongoing work, you may be acknowledged or included as a co-author in a future publication.

Entry requirements

You should have, or be studying, a degree in Mathematics, Biology or a related discipline. You should have a strong interest in genetics, genomics, or computational biology (although detailed knowledge of genomics or single-cell analysis is not required). Prior experience with programming or data analysis (for example, R, Python, or a similar language) would be advantageous. This project is ideal for students who are comfortable working with data and keen to apply or develop their coding skills in a biological research setting.

NDORMS 06
Developing an ultrasound-mediated nanoparticle drug delivery platform for treating bone metastases

Primary supervisor

Dr Dario Carugo

Project description

Bone metastases from a primary tumour are one of the most common sites of metastases, yet exhibiting high rates of mortality and morbidity. Patients suffer from cancer-induced bone pain, limited mobility, debilitating fractures, depression, cardiac arrhythmias and have an increased caregiving burden with reduced quality of life. Current methods of treatment are limited in improving outcomes and the lived experience of these patients.

The aim of the project is to investigate the efficacy of combining ultrasound-responsive nanoparticles with chemotherapeutic or targeted anticancer agents in vitro, using relevant bone-metastatic cancer cell lines. These particles start oscillating when exposed to ultrasound, which helps releasing a therapeutic agent at the desired location and point in time. Key techniques employed will include microbubble synthesis and characterisation, nanoparticle conjugation, particle sizing, microscopy, and cell culture–based cytotoxicity assays to identify drug targets. Fundamental data analysis and plotting will be utilised to assess efficacy.

Project outcome

You will gain practical, hands-on experience across both engineering and biological research disciplines. You will develop an understanding of nanoparticle and microbubble fabrication and characterisation techniques, as well as an understanding of mammalian cell culture and cytotoxicity assays to evaluate therapeutic efficacy. You will generate and analyse datasets evaluating whether the developed nanoparticle/microbubble formulations induce biological effects on cells. You will have the opportunity to present your findings to a multidisciplinary research group in an internal seminar and discuss your experimental results in the context of ongoing work. If your data or analysis contributes to a future publication or conference proceeding, you will be included as a named author.

Entry requirements

You should have, or be studying, a degree in a scientific or engineering discipline, such as Biology, Biomedical Sciences, Engineering, or a related discipline. Prior experience with, or understanding of, cell culture and basic microscopy techniques would be advantageous, but is not necessary. An interest in biomedical device development, cancer therapeutics, or drug delivery systems is desirable.

NDORMS 07
The role of vascular macrophages in health and disease

Primary supervisor

Professor Claudia Monaco

Project description

Macrophages have a key role in health and disease, by supporting organ function and immune responses to damage and disease. In homeostasis, ontogeny and organ-specific signals influence the phenotype and function of macrophages by activation of specific transcription factors. Using single cell (sc) RNA sequencing (RNAseq) and mass cytometry (CYTOF) in human and murine arteries, we revealed the existence of significant transcriptional and spatial heterogeneity of vascular and perivascular macrophages at the steady state. Distinct subsets are identifiable within the intima and adventitia of the arterial vessels. The functional and physiological relevance of this heterogeneity is unclear, but it points to different physiological functions related to immunity and metabolism. We developed genetically modified strains designed to evaluate the role of specific macrophage subsets in whole organisms and their origin, and state of the art imaging techniques to interrogate their intercellular interaction.

Project outcome

You will analyse data from disease relevant and combined models, human biobanks, combined with spatial proteomics technologies such as mass cytometry (and the relevant computational tools) to map the local interactions of vascular macrophages, immune and stromal cells and how they contribute to vascular health.

By completing this project, you will gain insight into the acquisition and analysis of single cell biology datasets from murine and human arteries and validation of key selective macrophage markers for identification of macrophage geography in arteries.

Entry requirements

This project, dependant on the successful candidate’s interests, may be computational or wet lab based and all necessary training will be provided. For a computer-based project, you should have, or be studying a degree in Computer Science or a related discipline. For a wet lab-based project, you should have, or be studying, a degree in Biology, Immunology, or a relevant discipline.

Paediatrics 01
Characterising vaccine-induced immune responses to outbreak pathogens

Project description

Measles outbreaks are tragic reminders that it is important to be vaccinated against measles to reduce the risk of infection, health complications, and death. Measles infects immune cells, predominantly B and T cells, and these cells themselves are subject to variation due to age or other infections such as HIV. It is of public health importance to advance understanding of factors influencing successful vaccination, eliciting a high level of measles-specific neutralizing antibodies.

This study will investigate the impact of age, HIV status, and immune cell counts on the outcomes of measles vaccination. In particular, this study will involve mathematical modelling and statistical analysis to estimate how age-dependent changes in the counts of different subsets of T and B cells influence the number and kinetics of measles-specific antibodies.

The results have the potential to inform decision-making for the administration of the first dose of measles-containing vaccines.

Project outcome

You will gain hands-on skills and coaching in scientific programming in R, as well as in the mathematical and statistical analysis of experimental and clinical data.

You will also gain knowledge of Good Clinical Practice (GCP), Information Security and Data Protection, as well as the standard operating procedures of the Oxford Vaccine Group, including those related to statistical analyses for clinical trials.

You will also have the opportunity to attend the weekly meeting of statisticians at the Oxford Vaccine Group, and present at the end of your internship. We would be pleased to include you as a co-author or acknowledge your contribution in a future publication, should any of your work with us be incorporated. You will also be invited to write a personal reflection on your experience, which may be published on the website or in the newsletter of the Oxford Vaccine Group.

Entry requirements

You should have knowledge or experience in at least one of the following areas: scientific programming (R, Python, Stata, SAS, Matlab, etc.), statistics, epidemiology, clinical trials, mathematics, machine learning, control engineering, bioinformatics, data science, infectious diseases, computational biology, vaccinology, immunology, biology, medicine, or any other relevant discipline.

Paediatrics 04
Mapping the molecular landscape of the hypothalamus in autism through single-cell multiomics

Primary supervisor

Professor Stephan Sanders

Project description

This project explores how gene activity and regulation differ in the hypothalamus between autistic and non-autistic individuals using single-nuclei multiomic data, where both RNA expression and chromatin accessibility (ATAC) are measured from the same nuclei. The dataset includes 40 human donors with whole-genome sequencing, allowing us to explore how genetic variation influences transcriptional and regulatory networks.

In this project, you will identify cell types and pathways showing altered transcription or regulation, and relate these to genetic risk for autism. You will compare your findings with bulk RNA-seq from the prefrontal cortex (PFC), single-cell data from other brain regions (PFC, putamen, and Wernicke’s area), and published datasets. Recent work has shown that autism-associated genes are highly expressed in the developing hypothalamus and thalamus; this study will test whether similar or distinct molecular patterns occur in the hypothalamus, offering new insight into how gene regulation in the brain may shape neurodevelopmental differences.

Project outcome

You will be trained in cutting-edge statistical, bioinformatic, and genetic techniques for analysing single-nucleus multiomic data, integrating RNA expression, chromatin accessibility, and genetic variation. You will gain experience using computational tools to explore how these molecular layers interact and differ across cell types in the human brain. The project offers the opportunity to work alongside leading experts in autism genetics and neurogenomics, within a research environment also focused on developing therapies for rare disorders. At the end of the project, you will present your findings to the group in an internal meeting. If your analysis contributes to future publications, you may be acknowledged as a named co-author.

Entry requirements

You should have, or be studying, a degree in Natural or Life Sciences, Statistics, Mathematics, Computer Science, or a related discipline. Familiarity with R and/or Python is required, and experience with data analysis or visualisation would be advantageous. An interest in genomics, neuroscience, or gene regulation would be beneficial, as the project involves interpreting biological data in a quantitative context. You should be motivated to learn new computational approaches, have good attention to detail, and be comfortable working independently as well as collaboratively within a research team.

Pathology 01
Mechanisms of protein degradation

Primary supervisor

Professor Pedro Carvalho

Project description

Project description

We study membrane proteins and how they control signalling and cellular responses to stress. These processes are implicated in multiple human diseases including cancer, neurodegeneration, inflammation and infection so, although we mostly do discovery science, our work has wide potential medical relevance, and we are also interested in the translational opportunities.

Our particular focus is the rhomboid-like superfamily. We were the first to discover rhomboids, and we proved that they were novel intramembrane proteases, conserved across evolution, and that they controlled growth factor signalling.

More recently we have become interested in the much wider superfamily of rhomboid-like proteins, the majority of which are not proteases. Of these non-protease rhomboid-like proteins, we especially focus on the iRhoms, which we discovered to be primary regulators of inflammation.

Our experimental approaches include genetics, cell biology, biochemistry and structural biology, mainly in mammalian cells but also with a variety of model systems.

Project outcome

You will gain experience with experimental cell and molecular biology, as well as participating in a research group focused on discovery science. By the end of the project, you should have completed some aspect of one of our projects. If what you do is included in a future publication you may be included as a co-author.

Entry requirements

You should have, or be studying, a mechanistic bioscience-related degree, such as Biochemistry, Biomedical Science, or Biology.

PCHS 01
How can cafeterias increase plant-based sales? An online study looking at different menu options

Primary supervisor

Dr Emma Garnett

Project description

Moving towards plant-based diets is essential to mitigate climate change and improve population health.

Our group’s previous research has found that people are more likely to choose plant-based options when these are 'unmatched' (eg falafel burgers vs chicken pies), rather than 'matched' (eg falafel burgers vs beef burgers) This could be because unmatched plant-based dishes have flavour and texture attributes that meat-eaters prefer (eg a meat-eater who dislikes pastry might choose a falafel burger over a chicken pie).

However, it can be challenging for canteens to serve unmatched meals which are extremely different and may have different sides (eg chips vs rice). Therefore, it is important to discover how 'unmatched' meals need to be to drive these effects. Eg is a korma sufficiently different from a Balti, or a lasagne from spaghetti?

In this internship you will help us set up an online randomised control trial to answer these questions.

University of Oxford: Pulse website
Cyclone Data Hub website
Neural Information Processing Systems website

Project outcome

You will gain hands-on experience working with real environmental and health datasets, learning how to process, visualize, and interpret geospatial data using platforms such as Google Earth Engine and Python. You will develop skills in linking environmental exposure data (industrial emissions, satellite imagery) with health outcomes (Acute Respiratory Infections in children) and in applying basic statistical and spatial analysis techniques.

By the end of the internship, you will produce a short data-driven report and visual summary identifying high-risk regions and their public health implications. You will also gain experience communicating scientific findings to non-technical audiences, and may be acknowledged in future publications or outreach materials based on this work.

Entry requirements

You should have, or be studying, a degree in Medicine, Public Health, Environmental Science, Data Science, Geography, or a related discipline. An interest in the intersection of health and environmental data is essential. Basic familiarity with data analysis (eg using Excel, R, or Python) and a willingness to learn simple statistical or geospatial techniques is desirable. No advanced coding experience is required, but curiosity about how data can inform public health and policy decisions will be important.

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

Population Health 02
Evaluating retinal language-image model (FLAIR) on UK Biobank colour fundus images

Primary supervisor

Professor Bartek Papiez

Project description

This project will evaluate the performance and interpretability of FLAIR, a large language–vision model, on colour fundus images from the UK Biobank. By comparing FLAIR’s disease predictions with existing clinical labels, the study aims to assess its potential for refining or supplementing sparse diagnostic codes for eye conditions such as glaucoma and diabetic retinopathy, as well as its zero-shot classification capabilities for systemic conditions like diabetes and hypertension.

The project may (subject to time constraints) also investigate whether FLAIR’s latent language representations can enhance clinical conditioning in generative vision architectures (eg diffusion models), enabling more morphologically and clinically accurate synthetic fundus images. Overall, this study will explore how multimodal AI can improve model control, clinical alignment, and the real-world utility of synthetic medical imaging data in research and diagnostics.

Project outcome

You will gain hands-on experience in AI model evaluation, medical image analysis, and the responsible application of advanced machine-learning methods to biomedical data. You will be trained in handling large-scale imaging datasets, curating and processing data, and benchmarking model performance using quantitative metrics. You will also learn how to interpret and visualise AI outputs in a clinical research context, contributing to open and reproducible practices in AI-driven healthcare research.

Project outcome

Our work is aiming to change this by increasing understanding of what GAD looks and feels like for young people, and using this knowledge to develop more effective psychological therapies for those who need them in the future. So far, we’ve found that young people with GAD find uncertainty particularly challenging, and that they use unhelpful coping strategies such as seeking reassurance, overpreparing for situations, and avoiding things that worry them.

It's really important that these findings (and others) are shared with young people and professionals in a clear and meaningful way. We are therefore seeking an intern to help us with a number of science communication tasks – such as creating short videos for social media and generating resources for clinicians. To make sure that output is relevant for its target audience, you will incorporate feedback from young people with lived experience of generalised anxiety and professionals who work with them.

Project outcome

You will have the opportunity to develop skills in science communication, learn the importance of Patient and Public Involvement (PPI) in research, and gain insight into mental health research. Creative and writing skills are key to this internship, as it involves translating scientific findings into accessible and engaging content and messaging. At the end of the project, you will have the opportunity to present your work to our team and for it to be disseminated more widely.

Entry requirements

You should have, or be studying, a degree in Psychology or a related discipline, and have a particular interest in mental health and science communication. Some basic knowledge about mental health would be helpful, but extensive background knowledge is not required. We’re looking for someone who is open to expanding their skillset and trying new things.

Psychology 03
Measuring toddler response styles when tackling tasks that aim to elicit executive functions

Primary supervisor

Dr Alexandra Hendry

Project description

Executive function difficulties are linked to poor mental and physical health, poorer academic outcomes, and lower quality of life, across a range of populations (Diamond, 2013; Snyder, Miyake and Hankin, 2015; Wallace et al., 2016). In this project you will work with video and performance score data collected from 200 toddlers with and without a family history of autism and ADHD to consider how response styles interact with cognitive and regulatory skills to shape behaviour. By reconceptualising how to measure executive functions in toddlerhood we may be able to better chart the development of these important skills. This project, as with all of the team's research, takes a neurodiversity-affirming approach.

Project outcome

You will be trained in behavioural coding, and basic statistical analysis. At the end of the project you will present your findings back to the group in an internal meeting. If any aspect of your analysis is included in a future publication, you may be included as a named co-author on that paper. You will have the opportunity to join lab meetings involving researchers engaged in both basic and applied science. In these meetings (and in additional one-to-one sessions if requested) we will discuss the 'hidden curriculum' of higher education and research, and explore ways to flourish in academia.

Entry requirements

You should have, or be studying, a Psychology related degree and have standard software development skills at an undergraduate level. Familiarity with R would be useful, but is not essential.

Psychology 04
Understanding the human visual system through high-resolution retinal imaging

Primary supervisor

Professor Hannah Smithson

Project description

The human retina captures light and converts it into electrical signals in the human central nervous system (CNS). It is a directly observable part of the CNS, providing a unique 'window' to the brain.

Research in the Oxford Perception Lab uses an adaptive optics scanning laser ophthalmoscope (AO-SLO) and an adaptive optics optical coherence tomography (AO-OCT) system developed in the lab to image the living human retina at single-cell resolution. We also develop computational techniques to better analyse and interpret the data or to theoretically understand the early visual system and neural processing in the retina.

This project will be supervised by Prof Hannah Smithson, Dr Jiahe Cui and Ms Sirui Liu. It will utilise advanced retinal imaging or computational techniques to deepen our understanding of the human visual system. Specific project aims will be defined before the start of the internship, including but not limited to, data collection (including human participant imaging), data analysis, and software development.

Project outcome

You will gain practical experience in cutting-edge retinal imaging technology, human psychophysical studies, data and image analysis algorithms, theoretical modelling techniques or machine learning. You will be immersed in an interdisciplinary environment working with both engineers and vision scientists and have opportunities to learn about a range of different research projects in the field of vision science.

You will join weekly lab meetings and be given the opportunity to present your own work at the end of the project. If an aspect of your work is included in a future publication, you may be included as a named co-author on that paper.

Entry requirements

You should have, or be studying a science degree. Our research is interdisciplinary, and we welcome students from a range of backgrounds, including Experimental Psychology, Neuroscience, Engineering, Physics, or Computer Science, with an interest in addressing visual neuroscience questions. A degree of familiarity with Python or Matlab programming would be useful.

Interdisciplinary Projects

Expand all

Projects advertised on this page are based in one of the University’s four academic divisions and are open to applications from a wide range of academic backgrounds:

Anthropology 01
Co-parenting among the chaos: what we can learn from parents with toddlers during the UK lockdown

Primary supervisor

Dr Paula Sheppard

Project description

Biparental care is a defining feature of our species; women are able to produce and raise multiple dependent children because they have help from others: partners, grandparents, even friends. However, we also have strong cultures around the division of labour: in contemporary capitalist economies, at least one, but often both, parents work outside of the home and domestic labour (household chores), including childcare, has to be negotiated between them. In the UK, one of the reasons that people have very few children nowadays is because co-parenting duties tend to fall more heavily on the mother and this makes parenting and work-life more incompatible for women. This gender inequality in the family can lead to resentment and ultimately to avoiding having more children.

This study will use interview data collected during the UK lockdown (2020) from mothers and fathers of toddlers to explore how co-parenting practices were affected during this time, and what this may teach us about parenting challenges today. The data will be thematically analysed and written up into an article for publication.

Project outcome

You will be trained in qualitative methods such as thematic and content analysis and will learn to use NVivo software. You will also gain experience working within a research team which requires good communication skills, responsibility, and time-keeping. You will be co-authors on the main output; an original research paper to be published in a peer-reviewed social science journal, which will be an excellent addition to the CV of an aspirant masters or doctoral student.

This project is interdisciplinary and covers topics in computer architecture, computer networks, software engineering, and digital equity. AI is an optional, relevant field of study for this work.

Project outcome

Entry requirements

Engineering 12
Artificial intelligence for cancer detection in medical imaging

Primary supervisor

Professor Jens Rittscher

Project description

Project outcome

Entry requirements

Funding information

This project may be funded by the Google DeepMind Research Ready Scheme.

English 01
Knotted Histories: Early modern global carpets, global exchange and the public country house

Primary supervisor

Professor Nandini Das

Project description

This project will involve working with the team on the 'Knotted Histories: Early Modern Global Carpets, Global Exchange and the Public Country House’ project, run between the Faculty of English at Oxford and the National Trust. This project aims to reconsider cultural and global contexts of early modern carpet production and their use beyond traditional approaches, focusing on craft makers and manufacturers alongside an examination of the cultural meaning of carpets.

As part of this, the project is focusing on a series of National Trust properties as Case Studies, considering both their historic and current carpet collections and the role carpets played in social spaces within these properties. This study is conducted through archival analysis, examining catalogues, wills, inventories, and other documentation as appropriate, as well as through material analysis, considering the physical qualities and designs of the carpets.

Project outcome

This project will use a simplified energy balance approach to identify prominent processes that influence changing land surface characteristics during the monsoon season in these state-of-the-art simulations, with an aim to advance the understanding of the major drivers of future droughts.

Project outcome

You will join the Climate Research Lab to work closely with a team analysing climate observations and models and will:

gain data analysis skills in Python;
gain knowledge of the latest cutting-edge atmospheric models;
gain skills working as part of a research team; and
gain scientific communication and presentation skills.

It is possible that your contributions lead to a short paper/conference presentation communicating the outcome of your internship research.

Entry requirements

You should have, or be studying, a degree in a relevant field, such as Physics, Earth Science, or Geography, although we would welcome students in Mathematics or Computer Science with an interest in weather and climate. You should have basic familiarity with software coding, such as Python or similar. You should be interested in atmospheric and climate science, but in-depth prior knowledge is not required.

Geography 04
Exploring observations and simulations of high-impact storms in Africa

Primary supervisor

Dr Francesca Morris

Project description

Mesoscale convective systems (MCSs) are thunderstorms which can last for hours or even days, and span hundreds of kilometres. Not only are they high-impact weather systems which contribute extreme winds and rainfall, but they can also modulate atmospheric circulations at scales beyond their own structures.

Few observations of MCSs in Africa exist, but the Oxford Climate Lab have recently led several field campaigns in Southern and Eastern Africa, during which MCSs passed through the field sites. Observations were regularly taken using radiosondes and lidar, and could provide a unique insight into the winds associated MCSs, and therefore their upscale effects on the atmosphere. This project will explore these observations and assess their potential to contribute to our understanding of the winds associated with MCSs, comparing these to state-of-the-art ensemble weather forecasts from the UK Met Office.

Project outcome

You will join the School of Geography and the Environment’s Climate Lab, a team of researchers studying weather and climate and specialising in the Southern Hemisphere. By the end of the project, you will have:

learned about global and regional meteorology and climate science;
learned how to analyse in-situ measurements from meteorological field campaigns;
gained understanding of numerical weather prediction models and their output;
gained advanced data analysis skills in Python, particularly for analysing geospatial data;
gained practical knowledge of applying advanced statistical techniques;
gained scientific communication and presentation skills; and
contributed to a short paper communicating the outcome of your internship research.

Entry requirements

You should have, or be studying for, a degree in a relevant discipline, such as Physics, Earth Science, or Geography. We would also welcome applicants with a background in Mathematics or Computer Science with an interest in weather and climate. Familiarity with basic programming in Python (or a similar language) and the ability to use a basic command line interface for computing is highly desirable, however, training in basic computing will be provided. An interest in atmospheric and climate science is helpful, but in-depth prior knowledge is not required.