The principal components (PCs) can be presented
graphically as a score plot. The first component of the PCA
explained approximately 29.1% of the variation, and the second
component explained an additional 14.1% of the variation. Although
the two components resolved the measured composition profiles of
samples grown in the Philippines and other regions, the PCA could
not distinguish between rice samples grown in Korea and China.
Thus, PLS-DA was carried out to enhance the poor separation obtained
with the PCA model (Fig. 2). PLS-DA is a projection method
that separates groups of observations by rotating the PCAs so that a
maximum separation among classes, i.e., geographical origin, is
obtained. As shown in Fig. 2B, good separation in the score plot of