Spatial modelling of soil microbial interactions and the emergence of purely spatial interactions

A large diversity of microorganisms lives in soils where they transform the available organic matter, and store or release into the atmosphere the carbon it contains. Individual cells of the same or different species interact together in metabolic networks, i.e. networks of interactions ranging from competition for a same resource to cooperation with an exchange of resources. Because soil is a very heterogeneous environment, these interactions are limited by the local presence of resources and species. Therefore, all the theoretically possible interactions are not realized in practice. Understanding the impact of spatial heterogeneity on soil metabolic networks is essential to improve our comprehension of the carbon cycle in soils. However, it remains very difficult today to study spatial heterogeneity and metabolic networks in situ.  
  
Here, we present a numerical model we developed to study the impact of microbial spatial distributions on metabolic networks. Our model is spatially explicit and individual based. Each cell has a spatially limited impact on its environment, in which it is able to take up some resources and transform them into other products, which are then released into the environment and can be used by other cells.  
  
In this work, we explore the emergence of a type of interaction that only arise when spatial heterogeneity is taken into account, the eclipse dilemma (a concept first developed in Metabolic Resource Allocation in Individual Microbes Determines Ecosystem Interactions and Spatial Dynamics, Harcombe et al., 2014): in some spatial configurations, two individuals competing for the same resource can eventually enter a cooperating dynamic by providing to a common partner species with which they exchange resources.  We have found that while competition for the same resource reduces the average amount of resource that each individual can obtain due to sharing, cooperation with a common partner can lead to a local increase in available resources that can exceed the effect of competition. Those local increases in variability of metabolic interactions showed that spatialization in soil models is indeed essential to a proper microbial representation in models.