The digital mouse: How data and models complement research

Nanomaterials can distribute differently throughout the body. With the help of a digital mouse, a computer-based animal model, researchers at Empa in St. Gallen are investigating which tissues these materials may reach.

October 13, 2026

What happens when a chemical or material enters our body? Where does it travel, where might it accumulate, and which organs might it affect? These questions are the starting point for the so-called digital mouse. It is not a virtual animal, but an AI-supported computer model.

 

Until now, such questions have been investigated in part through animal experiments. A substance is administered; after a certain period, the organs are analysed and the distribution of the particles is determined. The digital mouse takes a different approach. “The digital mouse combines existing in vitro data with models of distribution in the body and can thus estimate how a substance will be distributed in the body and which organs will be affected,” explains Prof. Dr Peter Wick, head of Empa’s “Nanomaterials in Health” research group.

 

Speed is a key advantage. A computer model can provide an early indication: What might work? Where are further investigations needed? Which hypothesis should be tested first? For this initial orientation, the model does not need to capture every biological complexity definitively. At the same time, the more reliable data are available and the better the model is trained, the more capable it can become.

From cell model to simulation

In vitro models study cells or tissues outside the body. Under controlled conditions, researchers can observe whether cells take up a nanomaterial, what reactions it triggers and whether unexpected effects occur. More complex systems, such as organ-on-a-chip models, can connect multiple cell assemblies and biological functions.

 

In silico models, by contrast, use data, calculations and simulations. Their predictive value therefore depends both on data quality and on the underlying assumptions. The digital mouse is not developed in isolation on a computer; it results from the interplay of different data sources and methods.

A contribution to the 3Rs

Depending on the research question, such models can contribute to Replace, Reduce and Refine. They may replace certain animal experiments, reduce the number of animals needed or help plan experiments more precisely. Whether such a contribution is possible must be assessed scientifically for the specific application.

 

The digital mouse therefore represents a research approach in which cell models, data analysis, computer simulations and—where still necessary—animal experiments are not seen as opposites. Its potential lies in combining them: providing faster orientation, enabling more targeted analysis and strengthening the scientific basis for the next steps.

References

Interview with Prof. Dr Peter Wick, Research group "Nanomaterials in health" at Empa in St. Gallen, 18th September 2026.

Image: Google DeepMind, adapted with Anthropic Opus 5.5.