Identification of barley and rye varieties using matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry with neural networks

Cereal varieties are normally identified using time-consuming methods such as visual examination of either the intact grain or one-dimensional electro phoretic patterns of the grain storage proteins. A fast method for identifi cation of wheat (Triticum aestivum L.) varieties has previously been develo ped, which combines analysis of alcohol-soluble wheat proteins (gliadins) u sing matrix-assisted laser desorption/ionisation time-of-flight mass spectr ometry with neural networks. Here we have applied the same method for the i dentification of both barley (Hordeum vulgare L.) and rye (Secale cereale L .) varieties. For barley, 95% of the mass spectra were correctly classified . This is an encouraging result, since in earlier experiments only a groupi ng into subsets of varieties was possible. However, the method was not usef ul in the classification of rye, due to the strong similarity between mass spectra of different varieties. Copyright (C) 2001 John Wiley & Sons, Ltd.