Internship - Machine Learning for Toxicity Classification Using Historical In Vivo Data

At Roche you can show up as yourself, embraced for the unique qualities you bring. Our culture encourages personal expression, open dialogue, and genuine connections, where you are valued, accepted and respected for who you are, allowing you to thrive both personally and professionally. This is how we aim to prevent, stop and cure diseases and ensure everyone has access to healthcare today and for generations to come. Join Roche, where every voice matters.

This internship supports the development of a Biological Read-Across (BioRA) framework aimed at predicting human-relevant organ toxicities by training machine learning classifiers on pre-labeled, harmonized in vivo data (e.g., histopathology, clinical chemistry, organ weights).

Working closely with experts in toxicology, data science, and systems biology, you’ll help build and validate predictive models that classify compounds by toxicity type, enabling mechanism-informed safety decisions early in drug development.

Join our team and work in close collaboration with (computational) toxicologists as well as other scientists, using state-of-the-art bioinformatics and biostatistics tools and methods and gaining toxicological insights from experts in the field.

• Data preparation: There is an established pipeline for rat data harmonization that needs to be implemented for the other species

• Model Development: Train and optimize ML classifiers (e.g. random forest, SVM, XGBoost, deep learning) to predict compound-induced organ toxicities based on existing labels and harmonized datasets. Evaluate model performance using metrics such as ROC-AUC, precision-recall, and confusion matrices.

• Mechanistic Insight Extraction: Analyze feature importance and explore biological interpretation of model outputs. Map model-informative patterns to known toxicity mechanisms using ontologies and prior annotations (e.g., AOPs) (optional)

• Workflow Automation: Package reproducible data science pipelines for classification and evaluation. Contribute to a scalable framework that can be used for future compound screening or model updates.

• Internal Communication: Prepare progress summaries and present results to stakeholders in Predictive Modeling, TSAC, and Clinical Safety teams. Support documentation for potential inclusion in publications or internal tools.

You are currently enrolled in a Master’s program or have completed your (Master’s or Bachelor) studies not longer than 12 months ago in bioinformatics, biomedicine or computational sciences with interest or working knowledge in biology, toxicology, and/or life sciences.

• Strong programming skills in Python;

• Knowledge and interest in pharmacology /toxicology/life sciences

• Strong interest in interdisciplinary modeling (biology × AI)

• Effective communicator and self-driven problem solver

• Good data engineering, visualization, and interpretation skills are desirable;

Non-EU/EFTA citizens must enclose a confirmation from the university that a compulsory internship is part of the training with their application documents.

Start: October 2025

Duration: 12 Months

Workload: 100%

A healthier future drives us to innovate. Together, more than 100’000 employees across the globe are dedicated to advance science, ensuring everyone has access to healthcare today and for generations to come. Our efforts result in more than 26 million people treated with our medicines and over 30 billion tests conducted using our Diagnostics products. We empower each other to explore new possibilities, foster creativity, and keep our ambitions high, so we can deliver life-changing healthcare solutions that make a global impact.

Let’s build a healthier future, together.