Figure 33. Factor spectrum of the first factor derived from the pyrolysis mass spectra of the set of DMD arid normal tissue samples mentioned in Tab.le 4. This main factor accounts for 59% of the total variance present in the data set. The intensities in this factor spectrum represent ai . ai -values (ai = loading of the mass intensity i to the factor; ai = SD of the mass intensity i), so that the spectrum shows that part of the measured variation (ai) on each of the mass peaks, which is described by the factor (ref. 101). Because the loadings ai can be either positive or negative (due to normalisation of the spectra to total intensity), the factor spectrum exhibits positive and negative series of intensities; both of these can be interpreted individually. The close resemblance of the positive and negative parts of this factor spectrum to the difference spectrum (Figure 32) should be noted. In the negative part of the factor spectrum three subsets of fragment peaks can be recognized more clearly than in the difference spectrum, viz. the protein series of peaks at m/z 34, 48, 117 and 131, the carbohydrate series at m/z 31, 32, 43, 60, 70, 72, 84, 85, 96 and 126, and the series at m/z 58, 59, 71 and 89, indicative of choline residues from phospholipids (see Figure 9). By factor rotation methods (see Section 6.7) these subsets of peaks could be optimized separately and thus have to represent different chemical components within the tissue materials. The factor scores of this first factor are given in the insert and represent the intensities (standardised values) at which this factor contributes to the average spectrum of each of the samples.
Was this article helpful?