Skip to main content


Table 4 Multiple independent logistic regression analyses of clinical/immunological data and clinical/immunological/methylation data reveal the latter produces stronger probability of predicting IR

From: Comparative DNA methylomic analyses reveal potential origins of novel epigenetic biomarkers of insulin resistance in monocytes from virally suppressed HIV-infected adults

  1. Separate logistic regression analyses were performed for clinical and immunological data (left panel) and a combination of clinical, immunological and epigenetic data (right panel). The strongest independent predictive models (AUC) for outcome (IR) for clinical/immunological data were either, fasting glucose, 120 min. OGTT, FRS, or total to HDL cholesterol ratio. The strongest independent predictors of IR for clinical/immunological/epigenetic data were the methylation states of four CpGs; cg27655935, cg02000426, cg10184328, or cg23085143. Significance at P < 0.05
  2. AUC area under the curve, OGTT oral glucose tolerance test