Classification of single- and double-mutant corn endosperm genotypes by near-infrared transmittance spectroscopy

A total of 1,176 grain samples representing 10 different single- and double -mutant genotypic classes of specialty starch corn were used for developing various classification models based on near-infrared transmittance spectra . The genotypes used included amylose-extender (ne), dull (du), sugary-2 (s u2), waxy (wx), ne wx, ne nu, ne su2, du wx and du su2. Two-class classific ation models (only two genotypes compared) were developed using partial lea st squares analysis (PLS) while three-way and multiclass models were examin ed using principal component analysis (PCA). The effectiveness of the calib rations was evaluated by examining the percentage of unknown grain samples incorrectly classified. In general, two-class models performed better than multiclass models. However, they did not show improvement when discriminati ng among genotypes with overlapping amylose contents such ae du vs. ne and ae su2 vs. ae. Three-way models including double-mutants and their correspo nding single-mutant counterparts had misclassification percentages typicall y <5% using 14 PCA factors but again, with the exception of models includin g genotypes with overlapping amylose contents such as ne du vs. ae vs. du. The best multiclass model using all 10 genotypic classes simultaneously rev ealed only two classes (ne su2 and du) with misclassification rates > 10% b ased on 16 PCA factors. This study demonstrates that, depending on the mate rial to be considered, near-infrared transmittance spectroscopy could be us eful when segregation of specialty starch hybrids grain from other grain ty pes is necessary.