Stratifying renal carcinoma patients using data-driven immune & epithelial landscapes

Located in the Gasthuisberg campus in the picturesque city of Leuven, the Computational Oncology team at KU Leuven (Department of Oncology) is at the interface of academic research and clinical practise. We are strongly interested in the use of machine intelligence to realise personalised health care. As such, our dedicated team of resarchers focuses on spatial and non-spatial omics, data integration and computer vision to power the next generation of biomarker detection strategies.

Project

Renal cell carcinoma (RCC) ranks in the top ten most frequent cancers and is responsible for approximately 134,000 deaths per year worldwide. Clear-cell RCC (ccRCC) accounts for 70 to 80% of cases and is characterized by genomic instability and dysregulation of the normal cell to cell communication in a spatially aware manner. Although ccRCC are known as immunogenic tumors, patient response has remained highly variable and no reliable molecular markers for patient treatment selection have entered the clinic thus far. Proposed biomarkers, such as tumor mutational burden, have consistently failed to predict ICB response.

To overcome this issue, several tumour classification methods have been proposed for renal cell carcinoma. Typically, these transcriptomics-driven approaches identify 4 to 6 distinct subtypes. However, these also fail to reliably predict therapy response. In this PhD position, you will build a novel classification system, using public and in-house datasets to improve upon existing methods. Key is a better precision and recall in identifying patients sensitive to therapy. To this end, you will work with different omics data types at single-cell and bulk resolution, while also extrapolating your findings in a spatial context. Subtasks in this project involve omics integration (information transfer), bulk sample deconvolution, summarization of functional modules and creation of classification models. All analyses will be rigorously benchmarked in comparison to existing patient subtyping strategies and reviewed by subject matter experts in a multidisciplinary manner.

Your mission

In this position, you will:

• Work with public and in-house single-cell and bulk omics data from patients.
• Design a data-driven cancer subtype classification approach.
• Benchmark the method in comparison to existing patient stratification approaches.
• Thoroughly test the reliability of the new classification system across patient cohorts.
• Explore how the signatures defining patient groups can help with future clinical trial design.

Expected results

• A cutting-edge, semi-supervised framework for subtyping of renal carcinoma patients that we have validated in several clinical trials.
• Machine-learning models with enhanced precision and recall to identify patients sensitive to immune checkpoint blockade.
• A set of cancer-related biomarkers that modulate therapy response and have prognostic value.

Profile

We are looking for highly motivated researchers who hold a Master of Science degree in either Bioinformatics, Biology, Biochemistry, Biomedical engineering or Medicine who either already have a strong affinity for machine learning or are motivated to pick it up. We welcome applications from individuals of any nationality and background, provided they meet the following requirements:

• Applicants must possess a MSc or an equivalent degree in a relevant field such as medicine, life sciences, engineering or a related discipline.
• Applicants may not hold a PhD degree or have successfully defended a PhD.
• Applicants must be eligible for enrollment in the PhD program at (KU Leuven).
• Applicants must be able to demonstrate proficiency in both written and spoken English (B2 level or equivalent).
• Applicants must have good analytical, writing, and presenting skills.
• Applicants must be eager to learn new things and have a team-oriented attitude.
• Experience with Python (especially PyTorch and scikit-learn) is a bonus.

Offer

We offer:

• A fully funded 4-year PhD scholarship with a competitive monthly stipend.
• Doctoral training to help develop your research career.
• Participation in groundbreaking interdisciplinary research at the interface between academia and the clinic.
• Opportunities to participate in national and international conferences.
• Access to state-of-the-art computing infrastructure and data acquisition technologies.
• Employee benefits (e.g. health insurance).
• A dynamic, passionate team of fellow PhD students, postdocs and clinicians.

Interested?

For more information please contact Prof. Stefan Naulaerts (stefan.naulaerts@kuleuven.be) or Prof. Benoit Beuselinck (benoit.beuselinck@uzleuven.be).

KU Leuven strives for an inclusive, respectful and socially safe environment. We embrace diversity among individuals and groups as an asset. Open dialogue and differences in perspective are essential for an ambitious research and educational environment. In our commitment to equal opportunity, we recognize the consequences of historical inequalities. We do not accept any form of discrimination based on, but not limited to, gender identity and expression, sexual orientation, age, ethnic or national background, skin colour, religious and philosophical diversity, neurodivergence, employment disability, health, or socioeconomic status. For questions about accessibility or support offered, we are happy to assist you at this email address.