A central result: cadherin-dependent adhesion contributes significantly to cell-cell tension in C. elegans, with a highly non-linear, Hill-like dependence. Adhesion is not just a small correction to contractility. 6/6

Inside the eggshell, the model predicts morphogenesis quantitatively: anterior compaction is driven by rising cell-medium tension, while P2’s early exclusion reflects high cell-cell contact tension. 5/6

To calibrate inferred tensions, we combined AFM measurements with cortical myosin intensity. AB-lineage cells are more tense than P-lineage cells, and tension rises dynamically through the cell cycle. 4/6

We combined live imaging, contact-angle measurements and tension inference to reconstruct relative surface tensions, then used a 3D active foam model to predict embryo shape over time. 3/6

The early C. elegans embryo is not mechanically uniform. Removing the eggshell, one can observe that at the 4-cell stage, ABa, ABp and EMS compact together, while P2 initially remains partly excluded — revealing dynamic, lineage-specific mechanics. 2/6

Very happy to have this work finally published! How do embryos sculpt their shape? We map the contractile & adhesive forces shaping early C. elegans embryos. With K. Yamamoto @Kazu_YAMAMOTO4 G. Charras' lab and my team shorturl.at/xv7G51/6

A true computational tour de force by Eric Neiva during his Marie Skłodowska-Curie postdoctoral fellowship in the team, now published in Journal Computational Physics: urlr.me/AuzK86 Congratulations, Eric! @CNRSbiologie Grateful to @EU_Commission for funding support

Beyond synthetic benchmarks, we show applications to mechanical parameter inference from microscopy images and to inverse design problems in epithelial tissues. 4/5

We benchmark three strategies for bilevel optimization in vertex models: automatic differentiation, implicit differentiation, and equilibrium propagation. This gives a practical comparison of their accuracy, speed, and memory trade-offs. 3/5

How can we learn tissue mechanics directly from patterns and images? In our new preprint, we introduce VertAX, a differentiable vertex-model framework in JAX for simulating epithelia, inferring parameters, and designing target tissue behaviors. shorturl.at/PUzT0 1/5