Skip to main content
. 2021 Oct 9;22(20):10903. doi: 10.3390/ijms222010903

Figure 3.

Figure 3

Principal component analysis (PCA) was used to derive the major distribution patterns from metabolomics data for all three groups. (a) PCA of 3749 metabolic features from 61 samples of S-MCI, P-MCI, and AD patients. (b,c) PCA representation of the twenty (b) or eight (c) metabolites selected by the machine learning algorithms. S-MCI, stable MCI; P-MCI, MCI proceeding to AD.