FRIPRO funding to develop multiscale cardiomyocyte models
Published:
Senior Research Scientist Karoline Horgmo Jæger has been awarded funding from FRIPROs Young Research Talent programme for her project, Combining spatiotemporal scales in computational models of cardiomyocytes avoiding averaging and homogenization (Scales).
The Research Council of Norway has awarded the project 8 million NOK. Running from 2026 to 2029, the project will involve a team of senior researchers and graduate students led by Karoline Horgmo Jæger.
Bridging the gap between scales
In medicine, symptoms appear at the organ level, but treatments often work on a molecular level. Understanding how these levels interact is critical for creating effective and safe treatments. However, standard computational approaches to link small and large scales—such as averaging or homogenization—tend to overlook important details.
This is the background behind the Scales project, aiming to address multiscale problems in computational physiology with an approach that avoids obscuring critical dynamics by homogenization or averaging.
Jæger and her colleagues will address this by replacing fine-scale dynamics with machine learning models, integrated into large scale-simulations while preserving the overall fine-scale resolution. This approach will drastically reduce the computational resources required to accurately integrate small-scale effects in models of large-scale phenomena.
Cardiomyocytes as a starting point
As an illustrative example, the project will focus on the computational analysis of cardiomyocytes, the muscle cells of the heart. For these cells a key challenge is understanding how tiny structures within and between cells influence their overall function.
Selected nanodomains– nanometer-sized volumes close to clusters of proteins– will be represented by highly precise machine learning models, enabling the inclusion of fine-scale dynamics in large-scale simulations of entire cells and collections of cells.
The systematic Scales approach and the models developed through this project could serve as a starting point for developing multiscale models across a range of physiological problems.
The Scales project will be part of Simula's scientific computing research efforts, collaborating with ongoing projects such as SUURPh and SIMBER. These partnerships involve leading institutions including the University of California, San Diego, the University of California, Berkeley, and Lawrence Berkeley National Laboratory.