Stable and unstable communities

A new mathematical model helps to understand how groups of microbes respond to changes in the environment.
  • Views 108
  • Annotations

Network of all possible regime shifts in the model, with 6 pairs of essential nutrients and 36 species. Image credit: Dubinkina, Fridman, Pandey and Maslov (CC BY 4.0)

In nature, different species of bacteria and fungi often live together in stable microbial communities. Exactly which species are present in the group and in which proportion may vary between communities. Changes in the environment, and in particular in the availability of nutrients, can trigger abrupt, extensive, and long-lasting changes in the composition of a community: these events are known as regime shifts. For instance, when bodies of water receive large quantities of phosphorus and nitrogen, certain algae can start to multiply uncontrollably and take over other species. A given community can have different stable species compositions, but it was unclear exactly how variations in nutrients can influence regime shifts.

To examine this problem, Dubinkina, Fridman, Pandey and Maslov harnessed mathematical techniques used in game theory and economics and modeled all the possible stable compositions of a community. They could then predict which environmental conditions – in this case, the amount of specific nutrients – were necessary for each stable composition to exist. These models also showed which conditions could trigger a regime shift. Finally, how resilient the communities were to different types of perturbations – for instance, an invasion by new species or changes in nutrient supply – was examined.

The results show that if competing species require different quantities of the same nutrients, then the community can have several possible stable compositions and it is more likely to go through regime shifts. In addition, a small number of keystone species were identified which can drive regime shifts by preventing other microbes from invading the community. Ultimately, these results suggest ways to control microbial communities in our environment, for example by manipulating nutrient supplies or introducing certain species at the right time. More work is needed however to verify the predictions of the model in real communities of microbes.