Skip to main content
. 2017 Dec 17;9(4):4522–4536. doi: 10.18632/oncotarget.23393

Figure 1. Multivariate statistics revealed the existence of a specific lipidomic signature of PCOS patients.

Figure 1

Both unsupervised Principal Component Analyses (PCA) (A) and supervised Partial Least Discriminant Analyses (s-PLS-DA) (B) indicate that it is possible to discriminate between healthy controls and PCOS patients basing on their plasma lipidome. (C) Cross validation (CV) analyses (10-fold CV method) indicates that we obtained the maximum accuracy using only one component. (D) Among the metabolites which most contribute to define the first component of PLS-DA we could identify (basing on exact mass, retention time, isotopic distribution and/or MSMS spectrum) the phosphatidylglyceride 33:0 (PG330), the phosphatidic acid 41:2 (PA412) and the ceramide t34:0 (Cert34:0). Unknown identities are represented as exact mass_retention time.