Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: PCA2GO: a new multivariate statistics based method to identify highly expressed GO-Terms

Figure 1

Bi-plot of fraction data. The first Principal Component was correlated with distinct PC variables (covariance matrix S). Therefore, the hierarchy becomes visible in plots of GO-terms of datasets, which were not highly affected by a characteristic value of the PC variable. The group of unspecific terms was strongly correlated with PC1 (Loadings are marked in red). The correlation of the PCA variable with a PC is visible and allows identification of PCA, which are connected to the PCA variable. Identified groups, which were connected to a PCA variable in the bi-plot, were coloured according to their subcellular localization.

Back to article page