Bioinformatic methods for stratification of obese patients and identification of cancer susceptibility biomarkers based on the analysis of the gut microbiome

Abstract

Obesity has an impact on health by increasing the risk of various diseases. However, these risks might also depend on the metabolic health status, as it seems that metabolically healthy obese subjects are under a reduced risk of suffering comorbidities such as colorectal cancer. The gut microbiome has an effect on obesity and metabolic disorders through several integration pathways, making it a potential therapeutic target for these diseases. In this study, we characterized the gut microbiota of 356 obese and non-obese European individuals with different comorbidities associated with obesity. Using approaches based on supervised machine learning and network biology, we found a set of biomarkers of interest for differentiating metabolically healthy from unhealthy subjects. Then, we performed a linear discriminant analysis of effect size on a population of 1593 colorectal cancer, adenoma and control subjects assembled by the COST Action ML4Microbiome to investigate their role in colorectal cancer risk. Four of our biomarkers appeared in both approaches, suggesting their possible role in colorectal cancer development, prognosis and follow up: Clostridium leptum, Gordonibacter pamelaeae, Eggerthella lenta and Collinsella intestinalis. Further research via longitudinal studies or experimental validation of these microbial species would be necessary to confirm this association.