Skip to main content
. 2022 Jan 23;12(2):104. doi: 10.3390/metabo12020104

Table 3.

Statistical parameters of the OPLS-DA and OPLS models derived from the 1H NMR spectra of urine samples. ASD, autism spectrum disorder; US, unaffected siblings; OPLS-DA, orthogonal projection to latent structure discriminant analysis; 1H NMR proton nuclear magnetic resonance.

OPLS-DA Model Permutation (400 Times) *
Component a R2X Cum b R2Y Cum c Q2 Cum d R2 Intercept Q2 Intercept
ASD vs. US 1P + 1O 0.229 0.801 0.504 0.394 −0.315
OPLS model
ASD children 1P + 1O 0.359 0.662 0.478 0.467 −0.331

a The number of predictive and orthogonal components used to create the statistical models. b, c R2X and R2Y indicated the cumulative explained fraction of the X block and Y block variation for the extracted components. d Q2 indicates cumulative predicted fraction of the variation of the Y block for the extracted components. * Q2 intercept less than 0.05 indicates a valid model.