Diffusion model predicts the geometry of actin cytoskeleton from cell morphology

Cells exhibit various morphological characteristics due to their physiological activities, and changes in cell morphology are inherently accompanied by the assembly and disassembly of the actin cytoskeleton. Stress fibers are a prominent component of the actin-based intracellular structure and are highly involved in numerous physiological processes, e.g., mechanotransduction and maintenance of cell morphology. Although it is widely accepted that variations in cell morphology interact with the distribution and localization of stress fibers, it remains unclear if there are underlying geometric principles between the cell morphology and actin cytoskeleton. Here, we present a machine learning system that uses the diffusion model to convert the cell shape to the distribution and alignment of stress fibers. By training with corresponding cell shape and stress fibers datasets, our system learns the conversion to generate the stress fiber images from its corresponding cell shape. The predicted stress fiber distribution agrees well with the experimental data. With this conversion relation, our system allows for performing virtual experiments that provide a visual map showing the probability of stress fiber distribution from the virtual cell shape. Our system potentially provides a powerful approach to seek further hidden geometric principles regarding how the configuration of subcellular structures is determined by the boundary of the cell structure; for example, we found that the stress fibers of cells with small aspect ratios tend to localize at the cell edge while cells with large aspect ratios have homogenous distributions.Author summary: We proposed a diffusion model-based machine learning system trained with cell outline and corresponding actin stress fibers data. The overall process can be divided into two steps. First, we cultured human fibroblast (HFF-1) cells and segmented cell outlines/stress fibers from microscopic images using a CNN (convolutional neural network) method and image processing techniques. Second, we trained the machine learning system with the diffusion model using sets of two corresponding images, the segmented actin stress fibers and the input images (cell outline). After adequate training, the network can transform from cell contour to its potential actin stress fibers distribution. This system can be well utilized for conducting virtual experiments predicting the distribution and localization of actin stress fibers by adjusting the input cell geometry. This system could provide a novel and experiment-free approach to quantify the stress fibers dynamics associated with the cellular morphology.