Imbalances in the human gut microbiome, or dysbioses, are associated with multiple diseases but remain poorly understood. Existing biomarkers of dysbiosis fail to capture the ecological mechanisms that differentiate healthy from diseased microbiomes. We have developed a metric, the ecological network balance index (ENBI), that quantifies the balance between positive and negative microbial interactions. This metric was inspired by the phenomenology observed in a model for gut microbiome dynamics that we introduce in this work, which revealed alternative stable states with distinct emergent microbial communities: a healthy state dominated by negative interactions and a dysbiotic state dominated by positive interactions. The ENBI robustly differentiates these states in both simulated and empirical datasets spanning multiple diseases and correlates with disease progression in conditions such as colorectal cancer, which underscores its potential as a diagnostic tool.
López et al. (Thu,) studied this question.