Meta-analysis of faecal microbiome studies followed by machine learning to identify intestinal disease-specific...

Alterations in the gut microbiome have been found to be associated with both inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS), suggesting that interaction with the microbiota might drive pathogenesis. Identifying microbial profiles for IBS and IBD will help in the development of treatment strategies tailored to each illness, particularly when coexisting symptoms make clinical management difficult. In this study a meta-analysis approach followed by machine learning was used for elucidating the differences in microbial composition and community dynamics in IBS and IBD respectively. Machine learning models developed using Random Forest classifier was able to achieve an accuracy of 98 % on training data and 100 % on blind dataset, and the distinctive role of Pseudomonas-associated microbial network, which also includes Streptococcus, Stenotrophomonas, Actinomyces and Intestinibacter, was clearly observed in IBS.