Next, PCA was used to derive NPs from brain nutrient concentrations. The first five NPs that explain the highest variance (each of which had an eigenvalue >1.0) among the 47 subjects are described in Table 3. No obvious outlier was detected in a PCA plot (data not shown). NP1 which has the highest variance accounted for 26.20% of the total variance. NP1 is described as higher concentrations of SFAs, MUFAs, and n-3 and n-6 PUFAs (all loading coefficients >0.40). NP2 is described by high concentrations of carotenoids (all loading coefficients ≥0.30), and NP3 is described by high concentrations of retinol, α-TP, and PK (all loading coefficients ≥0.40). The correlations of these nutrients in each NP are as depicted in Figure 1. The first five NPs accounted for 75.92% of the total variance of the original nutrient dataset.