Skip to main content
Fig. 2 | Cardiovascular Diabetology

Fig. 2

From: Machine-learning to stratify diabetic patients using novel cardiac biomarkers and integrative genomics

Fig. 2

Feature importance of physiological and biochemical characteristics from patients. a Using HbA1c for binary classification representing the factors positively (red) and negatively (blue) impacting the construction of the model, with size of the bars depicting importance. The b total nuclear methylation and c total nuclear hydroxymethylation of patients. SHAP binary depiction of the interaction between d total nuclear methylation and e total nuclear hydroxymethylation and HbA1c levels. f Not including HbA1c for binary classification representing the factors positively (red) and negatively (blue) impacting the construction of the model, with size of the bars depicting importance. SHAP binary depiction without HbA1c of the interaction between g total nuclear methylation and methyltransferase activity and h electron transport chain complex III and BMI. Examining the multiple classification effects of prediabetes, i A modified T-Plot where the main effects of biomarkers on the prediction output are shown along the diagonal axis whereas interaction effects are shown off the diagonal. SHAP depiction of patient separation with the individual and correlated effects of HbA1c and total nuclear methylation. SHAP multiple classification depiction of the interaction between j total nuclear methylation and HbA1c. SHAP values > 0.0 are diabetic (T2DM), SHAP values < 0.0 are non-diabetic (ND), SHAP values = 0 are either ND or T2DM without influence on the model. Groups are considered significantly different if P ≤ 0.05 = * compared to non-diabetic. All data are presented as the mean ± standard error of the mean (SEM). ND: non-diabetic; T2DM: type 2 diabetic; Nuc: nuclear; Mito: mitochondrial; 5mC: 5-methylcytosine; 5hmC: 5-hydroxymethylcytosine; HbA1c: glycated hemoglobin; binary: no diabetes and diabetes; multiple: no diabetes, prediabetes, and type 2 diabetes

Back to article page