Sparse species interactions reproduce abundance correlation patterns in microbial communities

Area of Science:

• Ecology
• Microbiology
• Computational Biology

Background:

• Macroecology has identified broad patterns in microbial community structure.
• Understanding the dynamical processes driving these patterns, particularly species abundance correlations, remains a challenge.

Purpose of the Study:

• To identify mechanisms explaining correlated species abundances in microbial communities.
• To develop population models that can reproduce observed correlation patterns.

Main Methods:

• The study proposes incorporating sparse species interactions into population models.
• A Bayesian inference algorithm was designed to extract interaction constants from empirical data.
• Lotka-Volterra constants were utilized as a successful modeling approach.

Main Results:

• The developed models successfully reproduced empirical probability distributions of pairwise correlations across diverse biomes.
• Inferred models also accurately predicted single-species macroecological patterns in abundance fluctuations.
• Analysis revealed that sparsity is a crucial feature of microbial interaction networks.

Conclusions:

• Sparse species interactions are a necessary mechanism to explain correlated abundances in microbial communities.
• The Bayesian inference approach provides insights into microbial interaction network properties.
• Sparsity is identified as a fundamental characteristic of microbial community dynamics.