PhD Position (all genders) in Computational Biology and Deep Learning for Spatial Omics

Main tasks

We are a newly funded junior research consortium (BMFTR programme "Zukunft eHealth") of computational scientists, pathologists and nephrologists at UKE Hamburg and RWTH Aachen University Hospital, with a partner at Universitas Mercatorum, Rome. Our goal is to use spatial transcriptomics, digital histopathology and clinical data to understand the molecular basis of glomerular kidney diseases.

The position is jointly led by Dr. Robin Khatri and Dr. Lucia Testa. Robin Khatri develops computational methods for single-cell and spatial omics and their application to immune-mediated kidney disease, with recent work published in Genome Biology (2024), Nucleic Acids Research (2026) and Bioinformatics (2025), and, together with clinical collaborators, in Nature Immunology (2025), Nature Medicine (2024), Nature Communications (2024) and Cell Reports (2026). Lucia Testa works on geometric and topological deep learning, including neural networks on simplicial and cell complexes, with contributions in IEEE Transactions on Signal and Information Processing over Networks (2024), the International Joint Conference on Neural Networks (2023), the ICML Topological Deep Learning Challenge (PMLR, 2023) and Scientific Data (2026).

The consortium is embedded in the Institute of Medical Systems Bioinformatics (Director: Prof. Dr. Stefan Bonn) and the Hamburg Center for Translational Immunology, with access to clinical expertise in nephrology and to the bAIome high-performance computing infrastructure. The PhD student will be based in Hamburg, with regular joint meetings with the partner group in Aachen.

Your tasks

You will develop computational and machine learning methods to analyse and integrate spatial transcriptomics data of different resolutions and technologies, to characterise tissue organisation, and to derive data-driven molecular heterogeneity of kidney disease and relate it to clinical measures. Depending on your background and interests, the focus of your project will lie either on the integration and analysis of spatial omics data or on deep learning methods that exploit the higher-order structure of tissue, with the two projects closely linked. Methods will be released as open-source software, applied to a well characterised cohort of human kidney biopsies, and evaluated together with our partners. You will present your results at international conferences and publish in peer-reviewed journals.

This position is a fixed-term position for 36 months at 65% of the regular weekly work hours and is to be filled as of January 1, 2027 (or at the earliest possible date).

Your Profile

• A fully funded PhD position in a newly established, BMFTR-funded junior consortium with a clearly defined research programme and dedicated resources for travel and conferences
• Work on cutting-edge AI methods with direct clinical relevance
• Access to large-scale multimodal biomedical datasets and modern GPU infrastructure
• Close collaboration with clinicians, biologists and AI researchers in a highly interdisciplinary environment
• Opportunity to publish in leading international scientific journals and conferences
• Structured doctoral training through the UKE graduate programmes
• Master's degree (or equivalent) in bioinformatics, computer science, computational biology, physics, mathematics, statistics or a related quantitative field
• Strong programming skills in Python and experience with the scientific Python ecosystem; experience with deep learning frameworks (PyTorch or similar) is required

Application

Applications should include a cover letter describing your motivation and relevant experience, a CV, academic transcripts, and contact details of one or two references. Links to code repositories or a thesis are welcome. Please indicate whether you would prefer a project focused on spatial omics integration or on geometric/topological deep learning; this is not binding.

• Solid foundation in statistics and machine learning, and interest in geometric or topological deep learning, graph neural networks or related methods
• Experience with single-cell or spatial transcriptomics data, or with computational pathology, is an advantage but not a requirement
• Interest in biomedical questions and the willingness to work closely with clinicians and experimental scientists
• Good written and spoken English; German is not required
• Independent, careful and collaborative way of working

Immunity status

Please note that employment is contingent upon proof of immunization or immunity against the measles virus, in accordance with applicable legal and medical requirements. Documentation (e.g., vaccination certificate) must be provided before employment begins.

Our Offer

• Fair and transparent compensation in accordance with our collective bargaining agreement ( TVöD/VKA ) (€ 63,600 – € 91,100), taking into account qualifications and professional experience. The salary range applies to a full-time position of 38.5 hours per week. The amounts listed are guidelines and do not constitute a salary guarantee.
• Targeted and individualized professional development at both the technical and project levels is provided, offering long-term prospects in a meaningful work environment
• Comprehensive continuing education and training programs at our UKE Academy for Education and Career
• Opportunities to help shape our “UKE INside” personnel policy through cross-functional and cross-hierarchical projects
• Sustainable commuting: Subsidies for the Deutschlandticket as a job ticket and Dr. Bike bicycle service
• Excellent health, prevention, and sports programs
• Family-friendly work environment: Partnerships for childcare, free vacation care, counseling for employees with family members requiring care
• A secure job, meaningful work, and a supportive team environment
• Structured onboarding and open knowledge sharing within the team
• Our employee restaurant offers a wide variety of culinary options; additional options are available at the “Health Kitchen” cafés and bistros and at a supermarket located directly on the premises