Skip to main content
. 2019 Aug 27;10:1965. doi: 10.3389/fimmu.2019.01965

Figure 3.

Figure 3

Plasma inflammation markers. (A) The random forest (RF) analysis of soluble factors resulted in predictive accuracies of 86.6% for HIVNC, Pre-ART, and ART. The soluble factors importance plots display the top 30 metabolites, which most strongly contribute to the groups' separation for HIVNC, Pre-ART, and ART. (B) Principal component analysis using PCAtool indicating grouping in different disease categories. (C) Hierarchical clustering analysis of ANOVA of top 20 differentially expressed proteins with false discovery rate (FDR) <0.001.