Table 3.

The determined signals/metabolites associated with BMI and HDP by linear regression modeling (n = 154).

Signal	Compound ID	Annotation/identification	Ontology	Odds ratio*	95%CI	FDR p value #
0.94_244.0926 m/z	6175	Cytidine	OL_1	2.22	1.41–3.50	0.032
14.51_631.3452 m/z	1114770-15-8	1-(9Z-Octadecenoyl)-sn-glycero-3-phospho-(1′-myo-inositol)	PDc	2.18	1.43–3.32	0.026
13.56_402.3130n	131769837	24-oxo-1alpha,25-dihydroxyvitamin D3 ++	PDd	2.34	1.51–3.62	0.025

^*Linear regression models adjusting for maternal age, whether smoked more than 100 cigarettes in their lifetime, and gestational week when serum samples and BMI were obtained were used to identify signals that were associated with BMI; Logistic regression models further adjusting for gravidity (frequency-matching variable) were used to identify BMI-associated signals that were associated with HDP.

^#False Discovery Rate (FDR) corrected p-value.

⁺⁺Annotation is made by matching against the public base with MS (PDd) and supported by the pathway enrichment results.