Fig. 1

Bagplots and gemplots under normaly distributed prinicipal components. Bagplots (left) and gemplots (right) for identifying outliers in the space of the first two or three principal components, respectively. The plots consist of inner bags (or inner gemstones) that contain 50% of the samples and an outer loop (or outer gemstone). Sample outside the loop or gemstones are flagged as outliers. While multiple experimental groups can be displayed in the same subspace of principal components, outlier detection can be performed separately for each group. Switching from two to three dimensions, new outliers can be found (sample 49) while other sample disappear as outliers (sample 64)