Fig. 1From: Unraveling the molecular heterogeneity in type 2 diabetes: a potential subtype discovery followed by metabolic modelingGraphical overview of the study design. This study included two supervised and unsupervised classification sections. At the supervised classification part, we used a machine learning approach to identify a set of genes whose expression patterns could discriminate T2DM individuals from healthy ones. At the unsupervised section, the clustering of T2DM patients was employed for potential subtyping of the diseaseBack to article page