Machine Learning Reveals Time-Varying Microbial Predictors with Complex Effects on Glucose Regulation

Abstract

The incidence of type 2 diabetes (T2D) has been increasing globally, and a growing body of evidence links type 2 diabetes with altered microbiota composition. Type 2 diabetes is preceded by a long prediabetic state characterized by changes in various metabolic parameters. We tested whether the gut microbiome could have predictive potential for T2D development during the healthy and prediabetic disease stages. We used prospective data of 608 well-phenotyped Finnish men collected from the population-based Metabolic Syndrome in Men (METSIM) study to build machine learning models for predicting continuous glucose and insulin measures in a shorter (1.5 year) and longer (4 year) period. Our results show that the inclusion of the gut microbiome improves prediction accuracy for modeling T2D-associated parameters such as glycosylated hemoglobin and insulin measures. We identified novel microbial biomarkers and described their effects on the predictions using interpretable machine learning techniques, which revealed complex linear and nonlinear associations. Additionally, the modeling strategy carried out allowed us to compare the stability of model performance and biomarker selection, also revealing differences in short-term and long-term predictions. The identified microbiome biomarkers provide a predictive measure for various metabolic traits related to T2D, thus providing an additional parameter for personal risk assessment. Our work also highlights the need for robust modeling strategies and the value of interpretable machine learning. IMPORTANCE Recent studies have shown a clear link between gut microbiota and type 2 diabetes. However, current results are based on cross-sectional studies that aim to determine the microbial dysbiosis when the disease is already prevalent. In order to consider the microbiome as a factor in disease risk assessment, prospective studies are needed. Our study is the first study that assesses the gut microbiome as a predictive measure for several type 2 diabetes-associated parameters in a longitudinal study setting. Our results revealed a number of novel microbial biomarkers that can improve the prediction accuracy for continuous insulin measures and glycosylated hemoglobin levels. These results make the prospect of using the microbiome in personalized medicine promising.

Association of gut microbiota with glycaemic traits and incident type 2 diabetes, and modulation by habitual diet: a population-based longitudinal cohort study in Chinese adults

“…We also confirmed several microbial genera–glycaemic trait associations that have been reported in a Finnish prospective study [ 8 ]. Specifically, in the Finnish study, Paraprevotella , [ Ruminococcus ] torques group and Family XIII AD3011 group were considered as the most predictive microbial biomarkers (three of the top five ranked) for type 2 diabetes-associated variables.…”

“…Several human studies have reported a cross-sectional association of the microbiota with type 2 diabetes [ 3 , 5 , 6 ]. Recently, two European cohorts with relatively moderate sample sizes ( n = 273 and 608, respectively) examined the prospective association of the gut microbiota with type 2 diabetes or glycaemic traits [ 7 , 8 ]. However, the results from these previous studies were inconsistent, and evidence from large prospective cohorts is still lacking.…”

“…Gut microbiota-targeted dietary intervention is a promising and cost-efficient method to reduce disease risk [ 12 , 13 ]. However, neither of the prospective studies [ 7 , 8 ] evaluated the association of dietary or lifestyle factors with gut microbial genera that were specifically associated with type 2 diabetes or glycaemic traits.…”

Dietary ε-Polylysine Affects on Gut Microbiota and Plasma Metabolites Profiling in Mice

“…In addition, ε-polylysine downregulated the obesity-regulating genera Collinsella , Negativibacillus , and Anaerotruncus as well ( 52 , 53 ). On the other hand, the abundance of Parabacteroides ( 54 ), Odoribacter ( 55 ), and Alistipes ( 56 ) were increased by ε-polylysine, a genus that played a positive regulatory role in lipid metabolism, and its content was negatively correlated with obesity and other diseases associated with fat metabolism. Akkermansia muciniphila can reduce the levels of relevant blood markers of inflammation and improve metabolic parameters ( 57 , 58 ).…”

Gut Microbiota and Targeted Biomarkers Analysis in Patients With Cognitive Impairment

“…In recent years, microbial marker severs as a non-invasive diagnosis tool for some diseases such as hepatocellular carcinoma, colorectal cancer, and type 2 diabetes (15)(16)(17). Gut microbial marker has the potential to predict the development of CI and may be an effective target tool to prevent CI in the elderly.…”