. Author manuscript; available in PMC: 2012 Jan 1.

Published in final edited form as: Clin Pharmacol Ther. 2010 Nov 24;89(1):97–104. doi: 10.1038/clpt.2010.250

Table 1.

Simple and partial regression coefficients for compounds in the “ nitrogen metabolism pathway” on changes in QIDS-C score after escitalopram therapy.

Name of Compounds	Simple Regression (p-value)	Partial Regression (p-value)
Glycine	2.71×10−⁵ (0.0054)	3.68×10−⁵ (<0.0001)

Glutamic acid	2.44×10−⁴ (0.22)	4.54×10−⁴ (0.0088)
Aspartic acid	−2.29×10−³ (0.069)	−2.41×10−³ (0.029)
Asparagine	−9.46×10−⁵ (0.70)	−6.04×10−⁴ (0.0057)
Hydroxylamine	−1.33×10−⁴ (0.0022)	−1.02×10−⁴ (0.0035)

Coefficients and p-values for simple and partial regression analyses of metabolite levels at zero time on escitalopram response are listed. A positive coefficient represents a negative association between levels of the compound and SSRI response, i.e., lower concentrations of metabolites are associated with better SSRI response. Glycine is “boxed” since it was the metabolomic “signature” pursued during the subsequent pharmacogenomic studies.