Figure 2. Discrimination between pre- and on-treatment serum samples.

(A) O-PLS models score plots for group T evaluating samples discrimination at W0 vs. W1 (1+1 components, R²X = 0.457, R²Y = 0.398, Q² = 0.061, CV-ANOVA p-value = 0.55) and W0 vs. W4 (1+1 components, R²X = 0.487, R²Y = 0.39, Q² = -0.230, CV-ANOVA p-value = 1). (B) O-PLS models score plots for group T+E, discriminating samples at W0 vs. W1 (1+1 components, R²X = 0.497, R²Y = 0.404, Q² = 0.199, CV-ANOVA p-value = 0.019) and W0 vs. W4 (1+2 components, R²X = 0.602, R²Y = 0.605, Q² = 0.301, CV-ANOVA p-value = 0.007). O-PLS model validations by re-sampling 1000 times the model under the null hypothesis for the treatment T+E. (C) and (D) O-PLS loadings plots represented for group T+E: W0 vs. W1 and W0 vs. W4, respectively. Statistically significant individual signals correspond to the colored spectral regions. Highlighted candidate markers are: 1) Acetate, 2) Acetoacetate, 3) Acetone, 4) Alanine, 5) Albumin Lysyl, 6) Betaine, 7) Choline, 8) Citrate, 9) Creatine, 10) Creatinine, 11) Fatty acids, 12) Fatty acids (mainly LDL), 13) Fatty acids (mainly VLDL), 14) Glucose, 15) Glycerol backbone of PGLYs and TAGs, 16) Glycerophosphocholine, 17) Histidine, 18) Lysine, 19) Methanol, 20) Myo-inositol, 21) Phenylalanine, 22) Tyrosine, 23) Valine.