Exploration and prediction of frozen surimi gel quality regulating factors based on LASSO regression

Abstract

Effect of freezing process parameters (cryoprotectants, freezing method, frozen storage temperature and time) on the gel quality of frozen surimi was evaluated and a predictive model was developed using least absolute shrinkage and selection operator (LASSO) regression method. Results revealed that the gel quality of the surimi decreased significantly with prolongation of frozen storage time. However, the selected cryoprotectants (silver carp hydrolysates, sericin peptides, and collagen peptides) effectively protected the textural and water-retention properties of frozen surimi gel. Specifically, samples treated with silver carp hydrolysates showed an approximately 29.5% higher gel strength and an over 4 % higher proportion of entrapped water (P₂₁) compared to the control group (p < 0.05). Liquid nitrogen freezing and frozen storage at −60 °C also effectively slowed down the gel degradation. LASSO regression analysis highlighted cryoprotectants (β₁ = 0.253) and frozen storage time (β₄ = −0.919) as key factors affecting the gel quality, with the model achieving a predictive performance of R² (0.877) and RMSE (0.341). This study offered new comprehensive strategies for processing and storage technologies of frozen surimi.