Figure 1

Mutations on the first and second principal components. All mutations having prevalences different from 0 are depicted. The wild type amino acid is cited before the codon of the mutation. Interpretation: PC1: First principal component (representing 11% of the variability), PC2: Second principal component (representing 6% of the variability). Mutations are represented by the component when they are close to the corresponding axis. When two mutations are far from the center, then, if they are: i) Close to each other, they are significantly positively correlated; ii) If they are in a rectangular position, they are not correlated; iii) If they are on the opposite side of the center, then they are negatively correlated. When the mutations are close to the center, it means that some information is carried on other axes.