Table 1 Comparison of model performances of PLS with sevenfold cross-validation between full metabolome vs post-selection.

From: Untargeted metabolomics reveals plasma metabolites predictive of ectopic fat in pancreas and liver as assessed by magnetic resonance imaging: the TOFI_Asia study

  

Full metabolome

Post-selection

 

nVar

nComp

R2Y

Q2

PCV-ANOVA

r

nVar

nComp

R2Y

Q2

PCV-ANOVA

r

Pancreatic fat

910

1

0.39

0.25

2.43E–04

0.63

56

3

0.81

0.69

2.12E–12

0.90

Liver fat

910

1

0.52

0.33

6.86E–06

0.72

64

3

0.80

0.66

6.11E–12

0.89

VAT/SAT

910

1

0.47

0.30

1.32E–05

0.69

31

2

0.70

0.62

1.88E–12

0.83

  1. nVar: number of variables; nComp: number of components; R2Y: goodness of model fit; Q2: predictivity; Pcv-anova: statistical significance of the PLS model; r: Pearson correlation coefficient.
Back to article page