The PMMH seminar of October 9 is postponed due to the closure of the campus.

Together with a model of the unsteady flow in the anisotropic, porous tip of the brush, these experiments allow us to identify different splatting regimes and to understand how splatter painting artists adjust their working conditions.

I present our efforts to unravel physical phenomena in splatter painting. We perform experiments on real and simplified brushes, subjected to a controlled impact on their handle. We study the dependence of the expelled liquid and the resulting patterns on the liquid viscosity and brush acceleration.

In this technique, perhaps most emblematically used by American painters Jackson Pollock (1912-1956) and Sam Francis (1923-1994), the brush never touches the canvas, and tuning the final pattern becomes a challenge for the artist.

In art, splatter painting consists in subjecting a paint-loaded brush to an acceleration (for example by “flicking” the brush or tapping its handle) to detach paint droplets or filaments. Those are projected onto the substrate to be painted, leaving filamentary or spotty patterns.

We discuss how interactions between active filaments lead to swarm formation and collective force generation. [1] M. R. Rashid, M. Akter, A. M. R. Kabir, K. Sada, T. Hiraiwa, A. Kuzuya, I. Kawamata, M. Tani*, and A. Kakugo*, ACS Nano 20, 18, 13586–13597 (2026).

The collective force increases linearly with the inferred number of kinesin motors over a wide range of swarm sizes, although the measured force is smaller than expected from a purely additive contribution of individual motors [1].

Here, we generate swarms of kinesin-driven microtubules (MTs), filamentous biopolymers driven by molecular motors, by introducing attractive DNA interactions between MTs. We quantify the collective force generated by the MT swarms using a custom-built electromagnetic tweezer setup.

Collective motion is widely observed in nature, from schools of fish to bird flocks and insect colonies, where groups can accomplish tasks beyond the capability of individuals. Such collective behavior raises an interesting question: can a group of active agents also generate force collectively?

References [1] A. Altieri, F. Roy, C. Cammarota, G. Biroli, Phys. Rev. Lett. 126, 258301 (2021) [2] J. Pasqualini, A. Maritan, A. Rinaldo, S. Facchin, E. V. Savarino, A. Altieri & S. Suweis, eLife 14, RP105948 (2026) [3] N. Patil, A. Altieri, arXiv:2607.24328 (2026)

Finally, I will discuss extensions beyond the standard logistic-growth and purely competitive frameworks that can promote coexistence, thereby offering a new perspective on the long-standing diversity–stability debate.

Through dataset analysis from both healthy individuals and patients with inflammatory bowel disease, I will show that distinct physiological states of the gut microbiome map onto different noise-driven and disorder-driven regimes of the Generalized Lotka-Volterra model.

I present a high-dimensional version of the Lotka–Volterra model incorporating demographic fluctuations and heterogeneous interactions among species. I will then explore its relevance to the human gut microbiome.

Inferring species interactions in highly heterogeneous ecological communities and developing a robust, predictive theory of ecological stability remain major challenges, due both to the amount of data required and the complexity of the underlying interaction network.