Analysis From the results of PCA, the conformational sampling is analyzed
on a 2D principal component plot as depicted in Fig.2.
Each point represents one conformation of the trajectory at a
timet. In general, only the first principal components (PC1, PC2,
and PC3) that account for approximately 60–70% of the total
variance of atom positional fluctuations, are analyzed. Indeed,
these first components capture the most significant global move-
ments, that is, the fluctuations of highest amplitude that are gener-
ally biologically relevant motions. On a 2D principal component
plot, the larger the cumulative variance on the two considered
principal components, the more meaningful is the distance between
the points. It means that similar conformations will be grouped
while diverse conformations will be separated on the 2D PCA plot.
Fig. 2PCA results for p38αconformational sampling. Individual maps: in green the 200 ns aMD trajectory, in
orange the 200 ns cMD trajectory and in blue the experimental structures. Scree plot: cumulative variance as a
function of the number of components
Enhanced Molecular Dynamics 415