Séminaire externe - Ada Altieri (Université Paris Cité)

Abstract: Accurately quantifying interactions within species-rich ecosystems poses significant theoretical challenges, necessitating advanced inference techniques and random matrix theory computations. I will first introduce a high-dimensional version of the Generalized Lotka-Volterra (GLV) model, which incorporates random species interactions and demographic stochasticity [1, 2]. Following this, I will propose a proof of concept aimed at capturing the complexity of the gut microbiota. By analysing metagenomic data from both healthy individuals and patients affected by inflammatory syndromes of the gastrointestinal tract, I will relate the different physiological states of the human gut microbiome to distinct noise-driven and disorder-driven regimes within the GLV framework [3].
Finally, I will discuss a generalization beyond the well-mixed approximation to investigate spatial effects in a metacommunity scenario. Varying the dispersal rate among spatial communities, together with demographic noise, can lead to the identification of either a second-order or a discontinuous phase transition with novel features [4].

[1] A. Altieri, F. Roy, C. Cammarota, G. Biroli, Phys. Rev. Lett. 126 (2021)
[2] G. Garcia Lorenzana, A. Altieri, Phys. Rev. E 105 (2022)
[3] J. Pasqualini, S. Facchin, E. V. Savarino, A. Maritan, A. Altieri* & S. Suweis*, to be submitted
[4] G. Garcia Lorenzana, A. Altieri* & G. Biroli*, PRX Life 2 (2024)