GC/MS based metabolomics: development of a data mining system for metabolite identification by using soft independent modeling of class analogy (SIMCA)

Table 4 Interclass distance resulting from SIMCA

Class name	Sugar phosphate	Organic acid	Sugar	Amine	Fatty acid	PC number	Important *m/z*
Sugar phosphate	0.00	1.21	1.05	1.85	1.79	1	89, 147, 217, 299
Organic acid	1.21	0.00	1.46	3.81	4.38	1	101, 133, 147
Sugar	1.05	1.46	0.00	2.72	2.53	1	89, 103, 147, 217
Amine	1.85	3.81	2.72	0.00	4.32	1	86, 100, 174
Fatty acid	1.79	4.38	2.53	4.32	0.00	1	117, 129, 132, 145

We used only one PC for all groups in order to make a robust model without over-fit. A distance close to zero indicates that the two classes are virtually identical, and the value above 1.0 indicates real differences. The important m/z contributed to a model was indicated, and the most important m/z was shown by bold type.

ISSN: 1471-2105