Table 3.
Metabolite features associated with CRC risk in conditional logistic regression-based analysis while controlling false discovery rate at 0.25
| Nominal p-valuea (FDR) | Odds ratiob (95% confidence interval) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Metabolite (ID levelc; Identifierd) |
LC–MS Mode | RT (s) | m/z Main fragment (additional fragments) | Univariable | Multivariablee | UV, earlier samplesf | MV, earlier samplesg | Univariable | Multivariablef | UV, earlier samplesf | MV, earlier samplesg |
| Unknown (5) | RP | 40.79 | 191.0403 (169.0584) | 0.0000 (0.0223) | 0.0000 (0.0970) | 0.1052 | 0.1500 | 0.7511 (0.6647–0.8488) | 0.7633 (0.6743–0.8640) | 0.8101 (0.6280–1.0451) | 0.8235 (0.6321–1.0728) |
| Unknown phosphate C3H9N2O5P (4) | RP | 40.61 | 207.0142 | 0.0000 (0.0792) | 0.0001 (0.2458) | 0.3782 | 0.3501 | 0.7420 (0.6447–0.8540) | 0.7535 (0.6535–0.8688) | 0.8718 (0.6426–1.1829) | 0.8579 (0.6220–1.1832) |
| Unknown (5) | RP | 407.11 | 696.2549 (697.2582) | 0.0001 (0.1481) | 0.0015 (0.4834) | 0.0382 | 0.0710 | 0.7639 (0.6653–0.8772) | 0.7930 (0.6871–0.9151) | 0.7032 (0.5041–0.9810) | 0.7254 (0.5119–1.0279) |
| Unknown (5) | RP | 40.99 | 151.0478 | 0.0002 (0.1481) | 0.0005 (0.4834) | 0.0959 | 0.1724 | 0.8043 (0.7182–0.9007) | 0.8150 (0.7262–0.9145) | 0.7904 (0.5992–1.0425) | 0.8207 (0.6178–1.0901) |
| Pyroglutamic acid (2; HMDB0000267) | RP | 40.75 | 130.0500 | 0.0002 (0.1481) | 0.0005 (0.4834) | 0.0621 | 0.1104 | 0.7903 (0.6990–0.8936) | 0.8025 (0.7087–0.9087) | 0.7488 (0.5526–1.0148) | 0.7751 (0.5669–1.0597) |
| Unknown (5) | RN | 329.25 | 310.8388 (308.8412, 310.8388) | 0.0002 (0.1481) | 0.0007 (0.4834) | 0.9240 | 0.8366 | 0.8107 (0.7257–0.9057) | 0.8246 (0.7372–0.9223) | 0.9883 (0.7753–1.2597) | 1.0279 (0.7916–1.3347) |
| Hydroxytigecycline (3; ChEBI ID 142709) | RP | 407.20 | 602.2791 | 0.0002 (0.1481) | 0.0012 (0.4834) | 0.0218 | 0.0504 | 0.7731 (0.6749–0.8857) | 0.7961 (0.6932–0.9143) | 0.6863 (0.4975–0.9466) | 0.7139 (0.5094–1.0006) |
LC–MS liquid chromatography-mass spectrometry. RT retention time. FDR false discovery rate. UV univariable. MV multivariable. RP reversed-phase positive. RP reversed-phase negative. ap-values for likelihood ratio test. bOdds ratios per one standard deviation increase in the exposure. cLevel of metabolite annotation according to the Schymanski scale [21]. dUnique identifier at HMDB, ChEBI, or PubChem. eAdjusted for potential confounders: body mass index, smoking status, education level, diabetes, alcohol intake, and recreational physical activity. fA subset of 374 participants (187 case–control pairs) had repeated sampling occasions, of which the sample collected closest to case diagnosis was used in the main analyses. The analysis showed here used the earlier sample from these participants. gAdjusted for potential confounders: body mass index, smoking status, education level, alcohol intake, and recreational physical activity (diabetes excluded from model due to only one sample with reported diabetes)