4D Cell Biology: Drug Discovery using Advanced Microscopy, Organoids, Digital Twins, and AI
We aim to revolutionize pharmaceutical research by integrating organoids, advanced microscopy, whole-cell modeling, and artificial intelligence into a novel pipeline for drug discovery and diagnostics. Mitochondria generate approximately 95% of the energy in the human body. What’s less widely known is that they form a dynamic, interconnected three-dimensional network that changes rapidly over time. This four-dimensional (4D: x, y, z, time) structure exhibits distinct phenotypes depending on cell type, cell fate, and exposure to pharmaceutical interventions.
In my talk, I will demonstrate how the 4D mitochondrial network can serve as a powerful biomarker—enabling us to predict cell identity, anticipate cell fate, and directly link drug action to cellular health.Our approach brings together four key technologies:
a)Human stem cell–derived epithelial organoids of the lung and intestine;
b)Lattice light-sheet microscopy (LLSM) to capture high-resolution 4D data from these organoids;
c) Whole-cell digital twins that use particle-based over damped Langevin dynamics simulations to model the mitochondrial network’s behavior and reproduce experimental 4D LLSM data; and
d)Self-supervised artificial intelligence to construct biologically interpretable latent spaces that relate mitochondrial phenotypes to drug responses.