Meta-analysis defines predominant shared microbial responses in various diseases and a specific inflammatory bowel disease signal

Background Gut microbial alteration is implicated in inflammatory bowel disease but is noted in other diseases. Systematic comparison to define similarities and specificities is hampered since most studies focus on a single disease. Results We develop a pipeline to compare between disease cohorts starting from the raw V4 16S amplicon sequence variants. Including 12,838 subjects, from 59 disease cohorts, we demonstrate a predominant shared signature across diseases, indicating a common bacterial response to different diseases. We show that classifiers trained on one disease cohort predict relatively well other diseases due to this shared signal, and hence, caution should be taken when using such classifiers in real-world scenarios, where diseases are intermixed. Based on this common signature across a large array of diseases, we develop a universal dysbiosis index that successfully differentiates between cases and controls across various diseases and can be used for prioritizing fecal donors and samples with lower disease probability. Finally, we identify a set of IBD-specific bacteria, which can direct mechanistic studies and design of IBD-specific microbial interventions. Conclusions A robust non-specific general response of the gut microbiome is detected in a large array of diseases. Disease classifiers may confuse between different diseases due to this shared microbial response. Our universal dysbiosis index can be used as a tool to prioritize fecal samples and donors. Finally, the IBD-specific taxa may indicate a more direct association to gut inflammation and disease pathogenesis, and those can be further used as biomarkers and as future targets for interventions.

Automated summary

This study reveals a common non-specific response of the gut microbiome in a wide array of diseases, which may cause confusion in disease classification. A universal dysbiosis index is developed to prioritize fecal samples and donors across different diseases. Additionally, specific bacteria associated with inflammatory bowel disease are identified as potential biomarkers and intervention targets.