The total ion chromatograms obtained from the seven groups of samples with five to six replicates each were analyzed in two steps. The first step involved automated data analysis by the XCMS program (Smith et al., 2006), which identifies, matches, and quantifies ions across all samples in the seven sample groups. As described previously (Vinayavekhin
and Saghatelian, 2009; Vi- nayavekhin et al., 2015), XCMS provided average mass-to-charge ratio
(m/z), retention time and integrated mass ion intensity (peak area; MSII) of each ion in each of the LC–MS chromatograms. In the second step, a pairwise comparison was performed be-tween the SHAM and OVX samples to identify ions associated with the estrogen-deficient conditions
induced by ovariectomy. Each ion in the XCMS output files was subjected to the following filters:
(i) statistical significance (t-test with p o 0.05), (ii) fold changes of Z 2, and (iii) a minimum
MSII of 5000 in the elevated samples (which was approximately at the limit of detection of the
instru- ments). Ions that passed these criteria were then inspected manually in the chromatograms
to remove isotopic ions or any false positives on the final list.