Chemometric Profiling Of Processing-Induced Nutrient Transformations In Sweet Potato (Ipomoea Batatas) And Sorghum (Sorghum Bicolor)
Okidhika, C. U.
Department of Chemistry, Faculty of Natural and Applied Sciences, Ignatius Ajuru University of Education, Rumuolemini, Port Harcourt, Rivers State, Nigeria
Okwelle, P. L.
Department of Chemistry, Faculty of Natural and Applied Sciences, Ignatius Ajuru University of Education, Rumuolemini, Port Harcourt, Rivers State, Nigeria
Keywords: Sweet potato, Sorghum, Nutrient transformation, Chemometrics, Principal component analysis, Hierarchical cluster analysis, Food processing, Proximate composition, Mineral content
Abstract
Sweet potato (Ipomoea batatas) and sorghum (Sorghum bicolor) are nutritionally significant staple crops widely consumed across sub-Saharan Africa and Asia. Despite their dietary importance, the effects of diverse processing methods on their multivariate nutrient profiles have not been systematically evaluated using advanced statistical frameworks. Conventional univariate analyses are insufficient to capture the complex and interdependent nature of processing-induced compositional changes, underscoring the need for a chemometric approach. This study aimed to employ chemometric profiling, specifically Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA), to comprehensively characterise and classify processing-induced transformations in the proximate composition and mineral content of sweet potato and sorghum subjected to boiling, drying, fermentation, frying, and microwaving, with unprocessed samples serving as controls. Fresh sweet potato tubers and sorghum grains obtained from farms in Rivers State, Nigeria, were processed using five standard treatments. Proximate composition (moisture, protein, fat, ash, fibre, and carbohydrate) was determined according to AOAC (2019) official methods. Mineral content (calcium, magnesium, potassium, iron, and zinc) was quantified by atomic absorption spectrophotometry and colorimetric methods following wet acid digestion. Multivariate datasets were subjected to PCA using a correlation matrix and to HCA using Ward's agglomeration method with squared Euclidean distance. All analyses were performed in triplicate and processed in SPSS version 25.0. Proximate analysis revealed that boiling elevated moisture in sweet potato to 79.60 ± 0.35% and in sorghum to 15.8 ± 0.4%, while drying and frying reduced moisture substantially. Protein was highest in dried sorghum (11.2 ± 0.4%) and fermented sweet potato (1.92 ± 0.05%). Fat content increased markedly in fried samples (sweet potato: 2.55 ± 0.08%; sorghum: 5.8 ± 0.2%). Carbohydrate content peaked in fried sweet potato (33.48 ± 0.85%) and dried sorghum (70.4 ± 1.3%). Mineral analysis showed that drying consistently retained the highest concentrations: potassium reached 368 mg/100 g in sweet potato and 325 mg/100 g in sorghum, while iron attained 1.20 mg/100 g and 2.25 mg/100 g, respectively. PCA on the full dataset demonstrated that PC1 and PC2 explained 57.3% and 28.6% of total variance (cumulative: 85.9%), with moisture and carbohydrate dominating PC1 loadings and fat and protein driving PC2 separation. HCA generated four distinct clusters: Cluster 1 (raw controls); Cluster 2 (boiled, highest moisture, lowest minerals); Cluster 3 (dried and fermented, concentrated nutrients, elevated minerals); and Cluster 4 (fried and microwaved, elevated fat and carbohydrates, moderate minerals).Chemometric profiling demonstrated that processing methods exert distinct and classifiable effects on the nutritional composition of sweet potato and sorghum. Drying and fermentation emerged as the most nutrient-preserving and mineral-concentrating methods, while boiling posed the greatest risk of mineral leaching. PCA and HCA together provide a robust, data-driven framework for optimising processing strategies to enhance the nutritional quality of staple crops, with practical implications for food scientists, nutritionists, and food system practitioners in food-insecure regions.