Table 2. Models used for prediction of incidence of type 2 diabetes with abbreviations and variables included in the models.

Model [ref]	Abbreviation	Variables
1: Non-invasive [2,17]	NI	waist circumference, alcohol consumption, smoking, systolic blood pressure, family history of T2D
2: Non-invasive+metabolomics	NIMet	AM + sorbitol, galacticol, mannose, galactose, uric acid, oxalic acid, glucaric acid-1,4-lactone, 3-methyl-2-oxopentanoic acid, 2-hydroxybutyric acid
3: Adiponectin [2]	AdM	serum adiponectin, HOMA IR, smoking, IGT, IFG
4: Adiponectin + metabolomics	AdMMet	AdM + sorbitol, galacticol, mannose, galactose, uric acid, oxalic acid, glucaric acid-1,4-lactone, 3-methyl-2-oxopentanoic acid, 2-hydroxybutyric acid
5: Metabolomics model	Met	sorbitol, galacticol, mannose, galactose, uric acid, oxalic acid, glucaric acid-1,4-lactone, 3-methyl-2-oxopentanoic acid, 2-hydroxybutyric acid
6: Optimized metabolomics	OMet	galacticol, mannose, galactose and 2-hydroxybutyric acid
7: Adiponectin without glucose measures	AdM2	serum adiponectin, HOMA IR, smoking
8: Optimized metabolomics with glucose measures	OMet2	IFG, IGT, galacticol, mannose, galactose and 2-hydroxybutyric acid