Partial least-squares regression study of the effects of wheat flour composition, protein and starch quality characteristics on oil content of steamed-and-fried instant noodles

Abstract

Steamed-and-fried instant noodles represent a fast growing product category; however, concern has been raised about high residual oil content by both consumers and manufacturers. To understand how wheat flour characteristics affect the oil content of instant noodles, samples of five Iranian hexaploid landraces, five US/Canadian samples, and seven local commercial wheat flours were analyzed for composition, protein and starch quality, and noodle-making quality. Partial least-squares regression (PLS) analysis between composition, protein quality properties and wheat starch gel properties (independent variables, X) and oil content of instant noodles (dependent variable, Y) resulted in a model that could explain 76.3% of the variability with a cross-validated correlation coefficient of 58.2%. PLS regression coefficient analysis showed that protein content and protein quality were the most important factors, and were negatively correlated with oil content of instant noodles. Starch quality properties, gel hardness, gumminess and chewiness as well as pasting properties were other important contributors. Starch gel hardness, gumminess and chewiness were positively correlated with oil content of instant noodles, while pasting properties were negatively correlated. © 2006 Elsevier Ltd. All rights reserved.