GMCVD: Detecting key modulating gut microbiota that contribute the cardiovascular disease risk toward personalized prevention

Abstract

Nowadays, cardiovascular disease (CVD) is one of the leading causes of death and disabilities worldwide. Atherosclerotic cardiovascular disease (ASCVD) is one of the most life-threatening subtypes of CVD. Recently, an increasing number of studies start to focus on the correlation and prediction of CVD based on the information of gut microbiome. In this study, by applying explanatory machine learning model, random forest-based computational pipeline called Gut Microbiome for CardioVascular Disease (GMCVD) was developed to conduct the ASCVD prediction and feature ranking. The top key several modulating gut microbiotas from genus and species levels Ire identified based on their strong contribution and correlation to the risk of CVD. These key disrupted microbiotas Ire validated by several external experimental studies, which demonstrate the reliability and efficiency of the machine learning based CVD risk prediction pipeline. With the detection of those specific modulating gut microbiotas, the personalized prevention method to reduce ASCVD risk using probiotics is provided based on varied microbiotas including Bifidobacterium, Clostridium, and Bacteroides. Therefore, GMCVD will facilitate the personalized prevention via gut microbiota to reduce the risk of cardiovascular disease.