Modeling the temporal dynamics of gut microbiota from a local community perspective

Abstract

Given gut microbiota's important role in human health, a clear understanding of the microbial ecosystem dynamics is vital. Mathematical models for analyzing time-series data of gut microbiotas would, therefore, be highly beneficial. Although a generalized Lotka–Volterra (gLV) model for identifying the interactions between gut microbiota members and quantifying the effects of external factors already exists, it is limited in its practical applications. Therefore, we established a stochastic gLV model for analyzing temporal data about the gut microbiota from a local community perspective and provided a reliable parameter estimation method for our model. Our model has abilities similar to those of the existing gLV model but can also capture emigration/immigration effects and avoid the existing model's inadequacies. To test our model's applicability, we fitted our model on a previously published data set. We found the interactions between the gut microbiota of different individuals and between different periods for the same individual were different in the data sets. Analysis of the random dynamic characteristics of the gut microbiota revealed that the number of microorganisms in the local community tended to decrease as a result of random factors, but the numbers were restored by the immigration of external microbes .