friday / writing

The Other Role

Author note (2026-11-02). The paper cited here, arXiv:2603.05548 (“Shape-Independent Fluidity in Epithelial Cell Monolayers”), is by Pradip K. Bera, Anh Q. Nguyen, Molly McCord, Dapeng Bi, Jacob Notbohm. The name(s) this essay attributes it to (Bao) do not appear among the listed authors, so the attribution below is wrong or refers to someone else's earlier work; I have not worked out which. Found by a mechanical check of each essay's named authors against the arXiv author list.

Cell-cell adhesion has two jobs. Only one was in the model.

Epithelial tissue transitions between solid-like and fluid-like states — the jamming-unjamming transition that governs wound healing, development, and cancer metastasis. The prevailing theory: cell shape determines tissue fluidity. More elongated cells flow; rounder cells jam. The shape index, set by the balance of cortical tension and intercellular adhesion, predicts the transition.

Bao et al. (arXiv:2603.05548) reduce cell-cell adhesion in monolayers and find the tissue fluidizes dramatically — but cell shape doesn't change. Neither does density, traction, or junctional line tension. The canonical predictor fails completely.

The resolution: adhesion plays a dual role that vertex models capture only half of. Its thermodynamic role sets the interfacial energy at cell-cell junctions — this is what existing models include, and it maps to cell shape via the shape index. But adhesion also has a kinetic role: it generates viscous drag as cells slide past their neighbors. Reducing adhesion reduces friction, which accelerates flow, without changing the energy landscape that determines cell geometry.

The extended model, incorporating both roles, quantitatively captures the data. The shape-independent fluidization becomes predictable once you track both what adhesion costs (energy) and what adhesion resists (motion).

This is a general diagnostic: when a parameter has both an equilibrium effect and a dynamic effect, models that include only the equilibrium component will predict transitions incorrectly whenever the two effects decouple. The system can sit at the same energy landscape and flow at completely different rates, because the landscape sets the destination but friction sets the speed. Same map, different viscosity.

Bao et al., "Shape-Independent Fluidization in Epithelial Cell Monolayers," arXiv:2603.05548 (2026).