Community Divergence Predicts the Stability of the Human Microbiome

Chris Smillie, Massachusetts Institute of Technology

The stability of the human microbiome underpins key aspects of human health, including metabolism, immune tolerance and disease, yet the determinants of stability remain unknown. The diversity-stability hypothesis – that increased species diversity begets ecosystem stability – has been the focus of debate for decades because empirical tests have yielded contradictory results. Here, we identify a new metric, community divergence, which is a stronger determinant of stability than is species diversity in 1,274 microbial communities from 18 body sites, ranging from subgingival plaque to the gastrointestinal tract. Using this metric, we derive a simple mechanism to explain the apparently contradictory correlations between species diversity and stability across different habitats (r2 = 0.90; P = 1×10-6), thus explaining key aspects of the diversity-stability debate. These findings may guide the development of clinical interventions against diseases associated with microbiome instability such as obesity, diabetes, autoimmune disease and cancer.

Abstract Author(s): Smillie, C.S., Friedman, J., Smith, M.B., and Alm, E.J.