Fig. 3From: Deregulation in adult IgA vasculitis skin as the basis for the discovery of novel serum biomarkersClassification of adult IgAV using Random Forest (a) Principal component analysis of serum analyte concentrations distinguished IgAV patients and HC. (b) ROC curve of Random forest algorithm with corresponding AUC (with 95% confidence interval). Red line presents ROC curve of a random classifier (AUC = 0.5). (C) First ten variables according to their relative importance for prediction of adult IgAV. IgAV, Immunoglobulin A vasculitis; HC, healthy controls; PC, principal component; ROC, Receiver operating characteristic; AUC, area under the curve; LBP, Lipopolysaccharide binding protein; ANGPTL4, angiopoietin-like 4; IL, interleukin; FABP4, fatty acid-binding protein 4; CCL, Chemokine (C-C motif) ligandBack to article page