Table 2.

Single metabolite associations with FDR-adjusted p-value < 0.05 in the discovery step.

Biochemical	Super-pathway	Sub-pathway	HMDB ID	HCHS/SOL			MESA
				OR	Unadj-p†	FDR p	Platform	OR	Unadj-p†	1-sided p	1-sided FDR p
Glutamate	Amino Acid	Glutamate Metabolism	HMDB00148	1.33	1.36E–04	0.014	Amide-neg	1.16	1.11E–02	0.006	0.019
Phenylalanine	Amino Acid	Phenylalanine Metabolism	HMDB00159	1.23	1.63E–03	0.049	HIL-pos	1.17	3.55E–03	0.002	0.014
Linoleoyl-linoleoyl-glycerol (18:2/18:2) [1]*	Lipid	Diacylglycerol	HMDB07248	1.25	8.27E–04	0.040	HIL-pos	1.11	3.47E–02	0.017	0.030
Oleoyl-linoleoyl-glycerol (18:1/18:2) [1]	Lipid	Diacylglycerol	HMDB07219	1.29	8.90E-05	0.014	HIL-pos	1.13	1.51E–02	0.008	0.019
1-palmitoleoylglycerol (16:1)*	Lipid	Monoacylglycerol	HMDB11565	1.21	1.15E–03	0.040	C8-pos	0.92	9.96E–02	0.950	0.950
1-stearoyl-2-linoleoyl-GPE (18:0/18:2)*	Lipid	Phosphatidylethanolamine (PE)	HMDB08994	1.25	9.89E–04	0.040	HIL-pos	1.03	5.37E–01	0.269	0.314
1-stearoyl-2-oleoyl-GPE (18:0/18:1)	Lipid	Phosphatidylethanolamine (PE)	HMDB08993	1.28	2.77E–04	0.019	C8-pos	1.04	4.49E–01	0.225	0.314

*Metabolite with * indicates they were identified based on accurate mass data, retention time and mass spectrometry but not reference standards. Therefore, the verification is not as robust as metabolites without *.

^†All results were adjusted for age, gender, BMI, study center, and Hispanic/Latino background in HCHS/SOL, and adjusted for age, gender, BMI, study site (site WFU and UCLA are combined due to low cell count), and race in MESA. Unadj-p is the abbreviations of raw p before the FDR correction. 1-sided p-values reflect the requirement of the same direction of associations in the discovery and validation datasets for concluding replication.

Significant values are in bold.