SPEC PerkinElmer 10.13
News|Videos|November 6, 2025

New Chemometric Analysis Explores How Soy Protein and Starch Interactions Shape Food Texture

Author(s)Will Wetzel
Fact checked by: Caroline Hroncich

A recent study published in the Journal of Chemometrics explored how chemometric modeling can be used to examine ingredient interactions in processed foods.

This video was made with the help of NotebookLM.

Processed foods are ubiquitous in society and in grocery stores. A recent study published in the Journal of Chemometrics explored how chemometric modeling can be used to examine ingredient interactions in processed foods (1).

The industrial production of food involves complex interactions between ingredients, especially within starch–protein systems (1,2). These interactions directly affect essential sensory attributes such as firmness, elasticity, and flow. To better understand these dynamics, the researchers systematically studied blends containing different ratios of soy protein isolate, gluten, and starch (1). They further compared the effects of three starch sources, wheat, corn, and cassava, on texture.

How did the researchers use texture analysis techniques?

The researchers utilized texture analysis techniques to measure flowability, penetration force, rupture force, deformability, and hardness of the prepared gels under controlled laboratory conditions. The results were then analyzed through chemometric modeling, employing linear, quadratic, and cubic polynomial fits to generate equations capable of describing each textural response surface (1).

To determine what mathematical models were best to capture the observed behaviors, the researchers conducted experimental tests, including analysis of variance (ANOVA). The findings of the study revealed that soy protein isolate had the most significant impact on textural properties, followed by gluten and starch (1). Moreover, the study highlighted strong correlations among the measured texture parameters and found notable similarities between corn and wheat starch behaviors, while cassava starch exhibited distinct characteristics (1).

According to the researchers, the results provide valuable insights for the food industry. By using chemometrics, researchers can predict and control the texture of food products, which is important for improving product quality and consumer acceptance (1). The integration of chemometrics into formulation studies not only enhances understanding of ingredient interactions but also supports the development of more consistent and high-performing food products.

The research underscores the importance of combining statistical modeling with experimental food science, which is an approach that is likely to shape future innovations in texture optimization across a wide range of processed foods.

References

  1. Lacerda, L. D.; Ziegler, V.; Winck, S.; da Silveira, N. P. Chemometrics Study to Understand the Interaction of Starch–Protein Mixtures and Food Texture. J. Chemom. 2025, 39 (11), e70081. DOI: 10.1002/cem.70081
  2. Zhang, J.; Liu, Y.; Wang, P.; et al. The Effect of Protein–Starch Interaction on the Structure and Properties of Starch, and Its Application in Flour Products. Foods 2025, 14 (5), 778. DOI: 10.3390/foods14050778

Related to this article

Sparks near Reno Nevada – Site of 2026 SciX Conference ©   DTM Media  -chronicles-stock.adobe.com
Wednesday's program at SciX 2026 kept returning to one question: can spectroscopy perform outside the controlled laboratory? Speakers took surface-enhanced Raman scattering (SERS), laser-induced breakdown spectroscopy (LIBS), mid-infrared photonics, chemometrics, and fluorescence excitation–emission matrix (EEM) spectroscopy into clinics, food plants, reactor coolant loops, brine fields, and drinking water supplies.
NIR Spectroscopy of the Brain and Other Complex Analyses, ©   Navaporn  -chronicles-stock.adobe.com
Near-infrared spectroscopy has quietly stopped being a niche calibration exercise and started acting like an autonomous diagnostic partner, reading consciousness in an injured brain, steering a fermenter without a human hand, and hunting invisible plastic in a glass of water. The tool chemists once used to check moisture in grain is now fusing with artificial intelligence to make decisions no spectrometer could make alone.
Infrared Reimagined: How FT-IR Spectroscopy Learned to Read Molecules, ©  Luis Eduardo  -chronicles-stock.adobe.com
FT-IR spectroscopy, long treated as a mature bench technique for confirming a carbonyl stretch or fingerprinting a polymer, has quietly become one of chemistry’s most versatile discovery engines: it now infers molecular structure straight from a spectrum without a reference library, resolves chemistry tens of nanometers wide, and screens a fingerstick of blood for disease in minutes. The last five years of published research show FT-IR moving from a confirmatory tool into a predictive, autonomous, and field-ready analytical platform.