A Computational Model of Interface Formation in Bacterial Colonies*

Bacterial colonies benefit from heterogeneity: cells differentiate into diverse physiology and gene expression states. During growth, these states form patterns. To uncover the functional relevance of emergent patterns, we must model how they arise from cellular growth, phenotype inheritance, and interactions. Here we present an agent-based model to predict patterns formed by motile and matrix-producing cells in growing Bacillus subtilis colonies. By incorporating phenotype inheritance, differential cell interactions, and escape of outer motile cells, our model predicts the emergence of a pattern: matrix surrounds a fractal-like motile population. We find that some properties of the motile-matrix pattern depend on the initial arrangement of cells, while others do not. Using box-counting, we show that the emergent interface exhibits a fractal dimension that increases as cells grow but eventually saturates as the thickness of the peripheral matrix layer exceeds the capacity of inner cells to push it away. The presence of the fractal interface correlates with larger colony growth rates and increases the proximity of motile and matrix cells, potentially promoting resource sharing. Our results illustrate how population-level properties emerge from the interactions of individual cells.