ABSTRACT Human lifestyle and physiological variables on human disease risk have been revealed to be mediated by gut microbiota. Low concordance between many case-control studies for detecting disease-associated microbe existed and it is likely due to the limited sample size and the population-wide bias in human lifestyle and physiological variables. To infer association between whole gut microbiota and diseases accurately, we propose to build machine learning models by including both human variables and gut microbiota based on the American Gut Project data, the largest known publicly available human gut bacterial microbiota dataset. When the model's performance with both gut microbiota and human variables is better than the model with just human variables, the independent association of gut microbiota with the disease will be confirmed. We found that gut microbes showed different association strengths with different diseases. Adding gut microbiota into human variables enhanced the association strengths with inflammatory bowel disease (IBD) and unhealthy status; showed no effect on association strengths with Diabetes and IBS; reduced the association strengths with small intestinal bacterial overgrowth, C. difficile infection, lactose intolerance, cardiovascular disease and mental disorders. Our results suggested that although gut microbiota was reported to be associated with many diseases, a considerable proportion of these associations may be spurious. We also proposed a list of microbes as biomarkers to classify IBD and unhealthy status, and validated them by reference to previously published research. IMPORTANCE we reexamined the association between gut microbiota and multiple diseases via machine learning models on a large-scale dataset, and by considering the effect of human variables ignored by previous studies, truly independent microbiota-disease associations were estimated. We found gut microbiota is associated independently with IBD and overall health of human, but more evidence is needed to judge associations between microbiota and other diseases. Further functional investigations of our reported disease-related microbes will improve understanding of the molecular mechanism of human diseases.

Determine independent gut microbiota-diseases association by eliminating the effects of human lifestyle factors