journal menu
![[Figure 2]](gm5051fig2mag.jpg)
|
|
Figure 2
Properties of the ICA analysis for GI. (a, b) Two plots of normalized contributions from GI data sets to the first (a) and second (b) ICA components, each for two target functions. The components' maps are normalized to a RMSD of 1 so the y-axis of the right-hand plot has a higher value due to the opposite signs of correlated contributors. (c) Scatter plot of ICA results for GI with kurtosis as the ICA-optimized target, projected onto the plane defined by the first and the second ICA components. The units are arbitrary because the method is scale-independent. However, their ratio carries information about the relative magnitude of the components. The colour gradient represents the density of the points. The plot shows symmetrical distribution along both axes, which indicates statistical independence between the first ICA component, representing temperature-independent radiation-induced structural changes in the GI structure, and the second ICA component, representing temperature-dependent radiation-induced structural rearrangements. (d) Number of peaks at different thresholds present in dρ/dD maps for each temperature separately and for the maps of two ICA components. |
![[Figure 2]](gm5051fig2.jpg)
 | JOURNAL OF SYNCHROTRON RADIATION |
ISSN: 1600-5775
