Figure 4.

Diagnostic outcomes for DKD and non-DKD in discovery phase and validation phase. (A, B) Predictive performance of microbial combinations including different key OTU number (computed by 5-fold cross-validation and random forest). POD index in discovery phase (C) or validation phase (E) was used to estimate diagnostic efficiency for DKD. Integrating fivefold cross-validation with random forest model, ten microbial OTUs was identified as optimal diagnostic biomarkers. (D) ROC curve of the DKD-related classifier using 10 microbial biomarkers to distinguish 120 DKDs from 232 non-DKDs. (F) AUC in validation phase verified the predictive power of diagnostic model in 60 DKDs and 116 non-DKDs. Corresponding 95% CI and P value were shown in graph. CV, cross-validation; POD, possibility of disease; ROC, receiving operational curve; AUC, area under the curve; CI, confidence intervals.