Table 5.

Site-specific validation methylation analyses based on discovery findings.

Metabolic syndrome status		Validation population: males and females (n = 166)			Validation population: males only (n = 100)				Validation population: females only (n = 66)
		CDH22 Chr20:44880277	CDH22 Chr20:44880264	CDH22 Chr20:44880250	CCDC8 Chr19:46915716	CCDC8 Chr19:46915706	CCDC8 Chr19:46915704	CCDC8 Chr19:46915701	MAP3K13 Chr3:185000790	MAP3K13 Chr3:185000779	MAP3K13 Chr3:185000774	MAP3K13 Chr3:185000760
Metabolic syndromea	Crude beta ± s.e. (p value)	1.034 ± 0.520 (p = 0.0488∗)	1.14 ± 0.508 (p = 0.264)	0.334 ± 0.488 (p = 0.5)	−0.149 ± 1.09 (p = 0.9)	0.200 ± 1.18 (p = 0.9)	0.355 ± 1.03 (p = 0.7)	−0.063 ± 1.09 (p = 0.9)	−0.584 ± 0.491 (p = 0.2)	−0.969 ± 0.367 (p = 0.0110∗)	−0.705 ± 0.481 (p = 0.1)	−0.696 ± 0.517 (p = 0.2)
	Adjusted beta ± s.e.	1.27 ± 0.531 (p = 0.0469∗)	1.29 ± 0.625 (p = 0.0414∗)	0.222 ± 0.646 (p = 0.7)	0.235 ± 1.20 (p = 0.8)	0.573 ± 1.31 (p = 0.7)	0.547 ± 1.12 (p = 0.6)	0.711 ± 1.15 (p = 0.5)	−0.258 ± 0.575 (p = 0.7)	−1.12 ± 0.437 (p = 0.0137∗)	−1.17 ± 0.578 (p = 0.0491∗)	−1.42 ± 0.583 (p = 0.0192∗)

Gives beta values with standard error for unadjusted and adjusted regression model with methylation site as dependent variable and metabolic syndrome status as the independent variable. CDH22 validation regressions performed in overall validation population, CCDC8 validation regressions performed in males only, and MAP3K13 validation regressions performed in females only. CDH22 regressions adjusted for age, gender, race, smoking status, and antipsychotic type. CCDC8 and MAP3K13 regressions adjusted for age, race, smoking status, and antipsychotic type. ^aReference group is subjects without metabolic syndrome; therefore, a positive beta value indicates hypermethylation in subjects with metabolic syndrome while a negative beta value indicates hypomethylation in subjects with metabolic syndrome. ∗ indicates statistical significance based on a p value below 0.05.