Retrospective Cohort Study
Copyright ©The Author(s) 2024.
World J Diabetes. May 15, 2024; 15(5): 898-913
Published online May 15, 2024. doi: 10.4239/wjd.v15.i5.898
Figure 3
Figure 3 Multivariate analysis of bile acids and unsaturated fatty acids in orthogonal partial least squares-discriminant analysis model. A: Orthogonal partial least squares-discriminant analysis (OPLS-DA) 3D model between type 2 diabetes (T2DM) with or without non-alcoholic fatty liver disease (NAFLD) groups. R2X = 0.227, R2Y = 0.054, and Q2Y = 0.026; B: The 200−permutation test demonstrated no overfitting in the OPLS-DA model [Q2 = (0, -0.025), R2 = (0, 0.028)]; C: The contribution of the metabolite to distinguish T2DM with or without NAFLD is indicated by variable influence on projection values. NAFLD: Non-alcoholic fatty liver disease; CDCA: Chenodeoxycholic acid; TLCA: Taurolithocholicacid; GDCA: Glycodeoxycholic acid; UDCA: Ursodeoxycholic acid.