Postdoctoral Researcher – Machine Learning

KU Leuven; Vlaams-Brabant; Sur place; EUR 60 000 - 90 000; Plein temps

Résumé du poste

KU Leuven invites a researcher to join its antiviral discovery team to develop and deploy ML models that analyze high-content imaging data to identify candidate antiviral compounds. You will work in an international environment within CAPS-IT and collaborate with biologists and chemists.

The role requires a PhD in ML/CS/bioinformatics and a proven publication record; excellent English communication and independence are essential. The contract is initially one year with possible extension.

Qualifications

• PhD in ML, CS, bioinformatics or equivalent.
• Strong analytical skills and ability to lead projects.
• Excellent English communication and publications.

Responsabilités

• Develop and optimize ML-driven models in antiviral screening.
• Extract phenotypic fingerprints at whole-well and single-cell levels.
• Cluster compounds and infer MoA; integrate toxicity to reduce false positives.
• Collaborate with biology and chemistry teams; integrate multi-omics data.

Connaissances

ML models

PyTorch/TensorFlow

Data cleaning

Data fusion

Imaging data

Chemo‑informatics

Statistical skills

Formation

PhD in ML/CS/Bioinformatics

Outils

CellProfiler

Description du poste

Responsibilities

Antiviral drugs are used to successfully treat infections such as with HIV and HCV. Yet, for most (life)-threatening and neglected infections, there are no such drugs. This leaves also critical gaps in epi- and pandemic preparedness. Antiviral drug discovery efforts typically focus on a few known targets. Yet, the biology of viral replication consists of many more complex processes that should harbor a wealth of undiscovered druggable targets. Thus, a large space of potential druggable biology is entirely ignored. We aim to fundamentally revolutionize antiviral target‑discovery by uncovering this terra incognita. To that end, we developed high‑throughput, multiplex, high‑content multiparametric phenotypic antiviral assays. These allow screening hundreds of thousands of molecules in our fully automated high biosafety screening facility CAPS‑IT against multiple viruses. You will be responsible for the development and deployment of advanced machine learning models that leverage the full complexity of the imaging data and that allow the selection of molecules that will serve for in‑depth virological studies. Ultimately, this will result in the establishment of the first‑of‑its‑kind “Atlas of Druggable Antiviral Targets”.

You will join a dynamic, multidisciplinary and international virology team with state‑of‑the‑art infrastructure, but will at the same time also be embedded in a team with extensive expertise in AI and machine learning for computational biology and chemo‑informatics. This will provide the opportunity to design novel machine learning approaches that leverage state‑of‑the‑art AI methods (deep learning, generative AI, Bayesian modelling, active learning, etc.) to combine cellular imaging data, chemical compound structure, viral genomes and other omics data.

You will take ownership of the implementation and optimisation of ML‑driven models in our antiviral screening pipeline thereby unlocking the full richness of the multi‑parametric data using advanced AI. You will extract and interpret fully detailed phenotypic fingerprints at whole‑well and single‑cell resolution in virus‑infected cell cultures. AI models will be used to cluster compounds and infer possible mechanisms of action, identify peculiar activity signatures and integrate cellular toxicity profiles to reduce false positives and guide compound prioritisation. The models will be populated and iteratively refined by converging evidence from downstream validation (such as chemo‑genetics, structural modelling, functional assays, thermal proteome profiling and omni‑omics), creating an adaptive and constantly evolving discovery pipeline. AI‑driven interpretation will exploit the full complexity of the dataset to expand the druggable antiviral target space.

Qualifications

We seek a researcher with strong machine learning modelling expertise and experience analysing challenging large‑scale data sets. Experience with cellular imaging data or virology/immunology is a plus.

You hold a PhD in machine learning, computer science, bioinformatics or an equivalent field. You combine strong analytical skills with the ability to work independently and lead collaborative efforts. You are a team player, proactive, solution‑oriented, and comfortable taking ownership of complex projects. Excellent English communication skills and a strong publication record are essential.

Relevant Skills

• Experience in creating and evaluating machine learning models.
• Familiarity with deep learning frameworks such as PyTorch or TensorFlow.
• Experience in data preparation, preferably in a bioinformatics context (data cleaning, filtering, etc.).
• Expertise in data fusion and relevant algorithms (deep learning, generative AI, kernel methods, Bayesian methods).
• Preferably, experience with high‑content imaging or cell imaging data (e.g., CellProfiler, convolutional neural networks).
• Knowledge of chemo‑informatics and drug discovery.
• Strong practical statistical skills (batch effects, confounders, experiment design).

Offer

We offer a fully funded position with a competitive salary in a friendly and stimulating environment within one of Europe’s most innovative universities. The contract is initially for one year, with possible extension after a positive evaluation.

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 [e-mail].