Figure 3: Principal component analysis model and receiver operating characteristic curve for colorectal cancer. (a) Principal component analysis (PCA) scores plot of colorectal cancer urine specimens from control specimens based on 12 marker metabolites. The PCA scores plot showed that different samples (normal group, colorectal cancer group including early stage group and advanced group) were scattered into different regions; (b) receiver operating characteristic (ROC) analysis was performed using the values determined by the first two principal components. Area under the curve (AUC) = 1.00.