Recent advances in sequencing technology paved the way for intense sampling of host-associated microbial communities. This large sample number in turn opens new opportunities to investigate microbial community dynamics. Human microbial communities, for instance in the gut and the vagina, can display sudden jumps in their community composition. It is currently unclear what drives these alterations in community composition. In general, community dynamics is influenced by a number of factors, including ecological interactions between community members and the history of the community. To study community dynamics, these factors can be encoded in mathematical models.
Here, I will present three models that can explain systematic differences in community composition observed in host-associated communities. In particular, I will discuss the role of multi-stability as a driver of alternative community types. Interestingly, all three models can generate gradients or clusters, depending on their parameter settings. In conclusion, the mechanisms behind alternative community types detected in human microbiota are currently not well explored and are a promising topic of future research.