Characterization of insulin microcrystals using powder diffraction and multivariate data analysis

Norrman, M.; Ståhl, K.; Schluckebier, G.; Al-Karadaghi, S.

doi:10.1107/S0021889806011058

Figure 6
Loading plot for the first principal component, PC1, of the principal-component analysis (orange). A loading plot describes the relationships between the observations (crystal samples) and the variables (intensities). A plot of the loading vectors indicates which of the variables are important, and corresponds to the directions of the samples in the score plot, i.e. variables with positive peaks in the loading plot are important for samples on the positive side of its component in a score plot. The major positive peaks in the loading plot are thus important for the samples on the right side of the score plot in Fig. 4

(a). Likewise, the negative peaks are important for the samples on the left part of the score plot. The powder pattern of the F crystals is coloured in blue. The normalized intensity of the F crystals is here put on a negative scale for clarity. The negative peaks of the loading plot coincide very strongly with the peaks of the F crystals, indicating that the PC1 is dominated by the F crystals. The positive loading peaks coincide with the peaks of the crystals located on the right side of the score plot in Fig. 4

(a) (D, J and B).

JOURNAL OF
APPLIED
CRYSTALLOGRAPHY

ISSN: 1600-5767

Volume 39| Part 3| May 2006| Pages 391-400

doi:10.1107/S0021889806011058

Search term		doi		Advanced search
Author		volume	page