Machine Learning Approaches for Modeling Complex Community Dynamics

Computational or mathematical modeling generally refers to a mechanism-based formalism that help us test hypotheses, expand our understanding of a system, or make mechanism-aware predictions. These models often vary in their level of granularity. For example, metabolic models focus on biochemical interactions on the level of enzymes and molecules while population models such as Lotka-Volterra focus on species-species interactions. What these models share in common is that for any moderately sized microbial community, they become exceedingly complex in terms of number of interactions and free parameters to evaluate. These modeling fields continue to advance rapidly to meet these challenges. In this talk, I will highlight the potential of different machine learning approaches to augment mechanistic modeling efforts and synergize with large data and modeling efforts in the microbiome sciences.

Learning Objectives:

1. What is a complex system?

2. Examples of modeling approaches for microbial populations.

3. Machine learning approaches for understanding higher-order concepts.