NMR pattern recognition of peracetylated mono- and oligosaccharide structures. Classification of residues using principal-component analysis, K-nearest neighbor analysis, and SIMCA class modeling

摘要: Abstract Attempts have been made to correlate the NMR chemical shifts of carbonyl carbons, anomeric protons, acetyl methyl and pyranose ring protons residue structures contained in peracetylated mono- oligosaccharides. The treated a comprehensive manner as components 13-dimensional vector characterizing each residue. Residues could then be classified according their structure glycosidic linkages neighboring residues using pattern recognition methods including principal-component analysis, K -nearest neighbor classification, SIMCA class modeling. Of these methods, soft modeling algorithm was most successful classifying group test structure, irrespective adjoining residues. chemical-shift variables used final those belonging C2 carbon, H2 H 1, H3, H4 were found important classes or discriminating between classes. In general, all support use an integrated approach treatment parameters applied structural analysis.