Presentation Information

[SS27-01]Data-driven inference of chaotic dynamics and network structure in intestinal bacterial communities

*Kei Tokita1 (1. Graduate School of Informatics, Nagoya University (Japan))

Keywords:

Microbial ecosystem,Parameter estimation,Lotka-Volterra equation,Replicator-mutator equation,Chaos,Rank-abundance relationship

Parameter estimation of intrinsic growth rates and inter-OTU (Operational Taxonomic Unit) interactions in microbial ecosystems has been conducted using OTU dynamics data based on a multi-species Lotka-Volterra equation[1]. The estimated inter-OTU interaction matrix consists of a random matrix, as hypothesized by May[2], with a small number of strong predatory interactions added. Long-term simulations of a 17-species Lotka-Volterra system using these estimated parameters reveal that 12 species go extinct, while 5 species coexist in a chaotic manner. Furthermore, simulations of the replicator-mutator equation[3] with the estimated parameters show that, depending on the mutation rate, OTU dynamics exhibit either convergence to a fixed point or chaotic behavior, including opportunistic extinction and resurgence[4]. Additionally, we report the emergence of a geometrically distributed OTU rank-abundance relationship[5] and the possibility that OTU groups connected by strong predatory interactions divide metabolic tasks, transferring intermediate metabolites along a supply-chain-like system.
----
[1] S. Marino, N. T. Baxter, G. B. Huffnagle, J. F. Petrosino and P. D. Schloss, ”Mathematical mod- eling of primaty succession of murine intestinal microbiota”, PNAS, 111, 439-444, 2014.
[2] R. M. May, “Will a Large Complex System be Stable?”, Nature, 238, 413-414 (1972).
[3] M. Nowak, Evolutionary Dynamics: Exploring the Equations of Life, Belknap Press (2006).
[4] K. Tokita, ”May’s dream: the interactions in biological networks were random matrices after all”, Proc. J. Conf. AROB-ISBC-SWARM 2023, 1398-1401 (2023).
[5] K. Tokita, et al, ICBP2023, Aug. 13-18 (2023). Seoul, Korea.