Growth-rate distributions of gut microbiota time series: neutral models and temporal dependence
E. BrigattiS. Azaele
E. BrigattiS. Azaele
Nov 2023
0被引用
0笔记
摘要原文
Logarithmic growth-rates are fundamental observables for describing ecological systems and the characterization of their distributions with analytical techniques can greatly improve their comprehension. Here a neutral model based on a stochastic differential equation with demographic noise, which presents a closed form for these distributions, is used to describe the population dynamics of microbiota. Results show that this model can successfully reproduce the log-growth rate distribution of the considered abundance time-series. More significantly, it predicts its temporal dependence, by reproducing its kurtosis evolution when the time lag $\tau$ is increased. Furthermore, its typical shape for large $\tau$ is assessed, verifying that the distribution variance does not diverge with $\tau$. The simulated processes generated by the calibrated stochastic equation and the analysis of each time-series, taken one by one, provided additional support for our approach. Alternatively, we tried to describe our dataset by using a logistic model with an environmental stochastic term. Analytical and numerical results show that this model is not suited for describing the leptokurtic log-growth rates distribution found in our data. These results effectively support a neutral model with demographic stochasticity for describing the growth-rate dynamics and the stationary abundance distribution of the considered microbiota. This suggests that there are no significant parametric demographic differences among the species, which can be statistically characterized by the same vital rates.
