The collision-induced dissociation (CID) fragmentation pathway data in Electrospray ionization mode coupled with principle component analysis (PCA) and Agglomerative Hierarchical Clustering (AHC) were used for structural analysis and classification of 3-oxosteroids.

All 3-oxosteroids analytes have been divided into 7 groups according to the number and position of double bonds and the presence of functional groups: 4-ene-3-keto; 1-ene-3-keto; 1,4-diene-3-keto; 4,9,11-triene-3-keto; poly-ene-3-keto; unconjugated 3-keto and hydroxylated 3-keto steroids. Tandem mass spectrometry measurements were used at the collision energy of 20eV to obtain product ions for each analyte. Experimental design and multivariate analysis were performed in XLSTAT (Addinsoft, NU, USA version 2009.3.02). All mass spectra data were converted into text files and then imported into XLSTAT as an array of m/z and intensity values. For all analytes the spectral data were referenced by finding out the common mass shifts due to the [M+H]+ ion. PCA and AHC analysis were applied to all previously referenced data transferred into XLSTAT matrix for the structural model evaluation. Analysis of the obtained model showed that the compounds were divided into 7 groups corresponding to the numbers and position of double bonds and the presence of functional groups.

Application of multivariate analysis to the ESI-MS/MS data provides to be potentially efficient method of structure analysis for 3-oxosteroids. The model was verified on the 3-oxosteroids analytes that were not previously used in model evaluation. The method outlined here can be used for the structural analysis of other steroids and other classes of analytes. This technique gives a powerful tool for the rapid identification of analytes in complex matrixes.