T-INSIL startup project
Digitial Twins are becoming essential to guide experiments and predict experimental outcomes. The advantage is double: by modeling, we understand better, while on the other hand running simulations on a computer are generally cheaper and faster than lab experiments.
Massive stem cell production, bio-printing, organ-on-chip: these are the technologies of tomorrow which will allow regenerative and personalized medicine to become reality. However, these technologies will require a lot of improvement and optimization before coming to full strength. People in labs will need to perform a lot of experiments with cell cultures requiring a lot of engineering before obtaining satisfying results. These experiments are often done by trial and error, which is costly and wasteful, and do not give the necessarily more insight.
Our software T-INSIL can run virtual cell experiments on a computer and analyse the dynamics of the system (see picture above). In our models, we represent every cell individually, along with its properties and variability. The laws of physics and chemistry guide us through the behavior of individual cells and their collective behavior. In a tissue, molecules are constantly transported in the extracellular space and over the cell membranes. We compute here the local concentrations of those molecules and the effects of reactions inside the cells which eventually determine the cell’s fate or behavior. In all the models, experimental images play a crucial role. We extract the maximum out of them to make sure that the models are well validated.
In this way, we are able to predict the growth of tissues based on what we know of individual cells. Our technology is the result of over more of 20 years of high quality academic research, where we have worked together with biologists and physicians. Our software can be regarded as the biotech analogue of -nowadays widely used- digital twins for designing in aeronautics and automotive industries. With our software, we can simulate living systems in combination with manufactured system which is strategic for example in the design of bio reactors.
Our expertise and partnerships
We have models that can reliably reconstruct a system of experiments or parts of an in-vivo system. These models are parametrized such that many virtual experiments can be generated, usually much quicker than the real experiment. Our model expertise includes:
- Toxicological pathways, predict how cells respond to a drug
- Metabolic pathways
- Tissue dynamics in response to external conditions (growth, homeostasis)
- Biomechanics, the way cells deal with mechanical forces
- Image processing (segmentation, analysis)
- Cell to cell variability, dealing with uncertainties
The user learns to understand the problem by systematically improve the models. Such a model can be very simple in the beginning, but the complexity will step-wise increased. The most important is that one grasps what the simulation results our telling us. Only then, we can keep makin progress.
We currently have a partnership with TreeFrog Therapeutics, a rising French biotech company who have developed a new technology to massively increase the production of stem cells. We are creating a digital twin for their system. T-INSIL have also been selected at the accelerator program of INRIA to launch this project.
If you are interested in a partnership to better understand and accelerate your system, please contact paul.vanliedekerke_at_inria.fr
external link:
