Skip to main content
. 2022 Feb 8;12:820240. doi: 10.3389/fphys.2021.820240

TABLE 3.

Receiver operating characteristic (ROC) curve analysis for individual biomarkers.

Metabolite AUC T-tests Log2 FC Cutoff Sensitivity Specificity LR + LR-
L-Threonine 0.9149(0.862–0.964) 1.1775E-13 –0.5072 6760000 0.7083 0.9792 34.0 0.2979
3-Methyl-2-butene-1-thiol 0.8594(0.768–0.922) 2.0004E-8 –0.4731 5780000 0.625 0.9792 30.0 0.3830
N-Acetyl-leukotriene E4 0.9523(0.909–0.98) 1.836E-21 1.9233 501000 0.9792 0.7917 4.7 0.0263
4-Hydroxy-6-docosanone 0.8889(0.829–0.943) 2.0327E-10 0.5019 535000 0.6667 0.9792 32.0 0.3404
LysoPC(18:0) 0.9566(0.921–0.983) 6.3462E-18 0.6343 17900000 0.98 0.8 4.7 0.0263
Dehydrophytosphingosine 0.9267(0.873–0.969) 1.0436E-10 –3.779 270000 0.7292 0.9792 35.0 0.2766
9-cis-Retinoic acid 0.9557(0.916–0.985) 0.001197 –8.4275 101.865 0.875 0.875 7.0 0.1429

AUC, area under the curve; FC, fold change; LR, likelihood ratio.