European Food Research and Technology最新文献

{"title":"Comparative performance of DNA and peptide-based methods for fish gelatin detection in commercial products","authors":"Adem Soycan,&nbsp;Mustafa Bilgin","doi":"10.1007/s00217-026-05301-7","DOIUrl":"10.1007/s00217-026-05301-7","url":null,"abstract":"<div><p>Gelatin is widely used in food, pharmaceutical, and nutraceutical products; however, extensive processing during gelatin production can compromise DNA integrity and complicate the detection of source-specific biomarkers. This study compared the performance of DNA-based real-time PCR (qPCR) and peptide-based Q-Exactive Orbitrap LC–MS/MS approaches for the detection of fish gelatin-derived biomarkers in commercial products and evaluated their performance across different product matrices. A total of 55 commercial products declared by their manufacturers to contain fish gelatin, including gelatin powders, capsules, tablets, and complex supplements, were analyzed using both methods. Fish DNA detection was performed by species-specific qPCR, while collagen-derived marker peptides generated by tryptic digestion were identified using high-resolution Q-Exactive Orbitrap LC–MS/MS based on accurate mass and MS/MS fragmentation criteria. The Q-Exactive Orbitrap LC–MS/MS approach detected the selected collagen-derived marker peptides in all analyzed samples (55/55), whereas qPCR detected fish DNA in 29 of 47 analytically valid samples (61.7%); eight additional samples were classified as IPC-invalid and were excluded from the valid qPCR analysis. The lower qPCR detection rate was observed particularly among capsule and tablet formulations, although the contribution of matrix effects, DNA integrity, and formulation-related factors could not be independently determined in the present study. No detectable fish-target signals or selected marker peptides were observed in the limited non-target control materials, including bovine gelatin, porcine gelatin, and gelatin-free plant-based matrices, under the applied experimental conditions. These findings indicate that, under the experimental conditions evaluated, peptide-based high-resolution mass spectrometry achieved a higher detection rate than DNA-based qPCR for the detection of the selected fish gelatin-derived biomarkers in commercial products. The study highlights the importance of selecting appropriate analytical targets according to matrix characteristics when evaluating fish gelatin-derived biomarkers in processed commercial products.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 10","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148903443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of antifungal adjunct cultures in plant-based cheese analogues","authors":"Zheng Zhao,&nbsp;Jinfeng Yin,&nbsp;Michael G. Gänzle","doi":"10.1007/s00217-026-05286-3","DOIUrl":"10.1007/s00217-026-05286-3","url":null,"abstract":"<div><p>The safety of plant-based dairy analogues remains poorly investigated. Antifungal adjunct cultures have been studied in dairy fermentations but their effect in plant-based systems remains unknown. In this study, antifungal adjunct cultures were evaluated in plant-based cheese analogues. Strains of <i>Lactiplantibacillus plantarum</i>, <i>Lacticaseibacillus paracasei</i>, <i>Furfurilactobacillus milii</i>, <i>Lentilactobacillus parabuchneri</i> and <i>Levilactobacillus hammesii</i> were selected to include both homo- and heterofermentative organisms. Each strain was inoculated into lupin- and buckwheat-based cheese analogues that were fermented with <i>Lactococcus lactis, Lactococcus cremoris</i> and <i>Leuconostoc pseudomesenteroides</i> as acidification cultures. Microbial community analysis by culture and full-length 16S rRNA sequencing revealed a strong matrix effect on the community composition. Lactate and acetate concentrations were higher in lupin-based products. Hydroxy-unsaturated fatty acids were identified both in lupin- and buckwheat-based products. Plant-based cheese analogues were challenged with environmental conidia, <i>Penicillium caseicolum</i> or <i>Penicillium roqueforti</i>. The adjunct cultures delayed growth of environmental conidia and of <i>P. caseicolum</i> in products with lupin and buckwheat. Growth of <i>P. roqueforti</i> 3969 was inhibited only in buckwheat products, resulting in a 4–10-fold extension of mould-free storage time. These findings demonstrate that antifungal adjunct cultures are also active in plant-based fermentations and emphasise the matrix effects of the antifungal activity of adjunct cultures.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decoding flavor differentiation in five representative dried chili peppers: an integrated sensory and multi-dimensional chemical profiling approach","authors":"Lili Zhao,&nbsp;Yiqing Li,&nbsp;Haobo Jin,&nbsp;Wei Zhou,&nbsp;Dongying Song,&nbsp;Mingqi Wang,&nbsp;Zuobing Xiao,&nbsp;Zhaogai Wang","doi":"10.1007/s00217-026-05294-3","DOIUrl":"10.1007/s00217-026-05294-3","url":null,"abstract":"<div><p>The condiment industry faces significant challenges with inconsistent chili quality due to variability in raw material. This study conducted a comprehensive analysis of color, pungency, taste and aroma compounds across five representative dried chili peppers (Qinjiao, Moguijiao, Sanyingjiao, Erjingtiao, Bohu). Integrated gas chromatography-ion migration spectrometer and gas chromatography-mass spectrometry analyses identified 18 key differential volatiles (variable importance in projection score &gt; 1) and 27 aroma-active compounds (odor activity value &gt; 1). Erjingtiao was the most sensorially preferred, which was strongly correlated with its exceptionally high linalool content (16-, 28-, and 5-fold greater than Sanyingjiao, Qinjiao, and Bohu, respectively), imparting dominant floral and citrus notes. Significant variations in capsaicinoid content were found, with Moguijiao possessing the highest level (6.49 g/kg), classifying it as very strongly pungent. Qinjiao and Sanyingjiao exhibited superior taste potential due to higher taste active value of umami- and sweet-associated free amino acids. Bohu displayed the most intense red color, as evidenced by its highest <i>L</i>*, <i>a</i>*, and color value (48.74, 26.36, 18.25). This study elucidates the distinct chemical-sensory profiles of major dried chili peppers, identifying key chemical markers responsible for their flavor differentiation. The findings are expected to provide scientific basis for a flavor-oriented raw material evaluation system, enabling the targeted selection and blending of chilies to optimize condiment quality and production efficiency.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Subsequent determination of added urea and solid non-fat (SNF) in cow milk: a rapid and non-destructive ATR-FTIR strategy using DD-SIMCA and OPLS","authors":"Fairuz Islam,&nbsp;Md. Athar Ishtiyaq,&nbsp;Md. Mehedi Hasan,&nbsp;Mst. Nusrat Jahan,&nbsp;Mahmudul Bhuiyan,&nbsp;Md. Azmain Faique,&nbsp;Md. Mahasin Ali,&nbsp;Md. Najim Uddin,&nbsp;Mohammad Hassan Shahavi,&nbsp;Md. Atikul Islam","doi":"10.1007/s00217-026-05290-7","DOIUrl":"10.1007/s00217-026-05290-7","url":null,"abstract":"<div><p>This study aims to develop a rapid, non-destructive technique using Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) with Data-Driven Soft Independent Modeling of Class Analogy (DD-SIMCA) and Orthogonal Partial Least Squares (OPLS) for the subsequent determination of added urea and Solid Non-Fat (SNF) in cow milk. For this purpose, different sets of adulterated milk samples were prepared by adding water (2.50–50.00% <i>v/v</i>) and urea (0.14-5.00% <i>w/v</i>) in various concentrations. Thereafter, all samples’ spectra were acquired over the range of 4000 to 400 cm^− 1 using an ATR-FTIR spectrometer for further multivariate analysis. Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) showed distinct clustering between control and urea adulterated milk samples in the amide I (1690 –1575 cm^− 1) region, while PC1 and PC2 explained 99.51% of the dataset’s variability. The DD-SIMCA model, designed for added urea detection, achieved 96.87% classification accuracy for the train and test sets. The optimum OPLS models with <i>k</i>-fold cross validation exhibited strong quantification of urea and SNF, achieving the highest calibration and validation performance with low prediction errors and sensitivity limits (model C, urea: R²X = 0.991, Q² = 0.997, RMSEP = 0.058, LOD/LOQ = 0.084/0.277; model Q, SNF: R²X = 0.998, Q² = 0.967, RMSEP = 0.482, LOD/LOQ = 0.501/1.519). Finally, ATR-FTIR spectroscopy with DD-SIMCA and OPLS techniques provides a robust, efficient alternative to traditional analytical methods for subsequent quality control of added urea and SNF content in milk for the dairy industry.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In vitro and in silico investigation of Bingöl honey, the first and only Turkish polyfloral honey with European union geographical indication: comprehensive chemical characteristics and biological activities","authors":"Ebubekir İzol","doi":"10.1007/s00217-026-05265-8","DOIUrl":"10.1007/s00217-026-05265-8","url":null,"abstract":"<div><p>Honey is a natural product valued for its nutritional and therapeutic properties. This study comprehensively investigated the chemical composition, quality parameters, antioxidant capacity, and enzyme inhibitory activities of Bingöl honey, produced in Bingöl and recognized as the first and only polyfloral honey with geographical indication registration from the European Union. Chemical parameters (aroma profile, mineral composition, moisture (12.6%), pH (3.82), electrical conductivity (0.4859 mS/cm), free acidity (37.21 meq/kg), proline (863.2 mg/kg), diastase (37.4), HMF (4.2 mg/kg), sugar profile, and antibiotic residues) were determined using validated analytical methods. Advanced techniques including UHPLC-Orbitrap-HRMS (major compound 3-(4-hydroxyphenyl) propionic acid (19,264.28 µg/kg)), LC–MS/MS (91 components were not detected), ICP-MS (the highest K (455.1 mg/kg)), HPLC-RID/DAD, GC–MS/SPME/HS, and affinity chromatography were employed for detailed chemical profiling. Total phenolic and flavonoid contents were quantified, and antioxidant activity was evaluated using DPPH (IC₅₀: 39.8), ABTS (IC₅₀: 41.8), and DMPD (IC₅₀: 48.9 µg/mL) radical scavenging assays and Fe³⁺ (1.212), Cu²⁺ (1.059), and FRAP (1.005 µg/mL) reducing power methods. Enzyme inhibitory effects against hCA I (11.35), hCA II (13.6), AChE (10.33), BChE (6.11), α-glycosidase (12.96), and α-amylase (13.84 µg/mL) were assessed by calculating IC₅₀ values. Molecular docking studies were conducted using six enzymes and the secondary metabolites present at the highest concentrations. Results indicate that Bingöl honey exhibits a rich phytochemical profile, significant antioxidant activity, and moderate inhibitory effects on multiple enzymes, suggesting its potential as a functional food and a natural source of bioactive compounds for the management of oxidative stress and metabolic disorders.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of novel functional fermented goji-based beverages using lactic acid bacteria and yeasts","authors":"Francisco Pérez Nevado,&nbsp;Rubén Magro Rubio,&nbsp;Mª Belén Ramos Alguijo,&nbsp;J. Carlos Solomando,&nbsp;Mª José Rodríguez Gómez,&nbsp;Patricia Calvo Magro","doi":"10.1007/s00217-026-05240-3","DOIUrl":"10.1007/s00217-026-05240-3","url":null,"abstract":"<div><p>Goji berry fruits (<i>Lycium barbarum</i> L.) are recognized for their health benefits due to the presence of bioactive compounds such as organic acids, polyphenols, and carotenoids. The aim of this study was to develop and characterize novel functional fermented goji berry juice beverages with lactic acid bacteria (<i>Lactiplantibacillus plantarum</i>), and yeast (<i>Saccharomyces cerevisae</i>), applied in both monoculture and co-culture. All tested strains exhibited good growth in goji juice, maintaining high viability throughout the 15-day fermentation period, with microbial counts remaining above 7 mL^− 1 for yeasts and 8 log CFU mL^− 1 for lactic acid bacteria. While <i>L. plantarum</i> led to a significant increase in acidity and phenolic content (0.32–0.79% and 4.84–5.77 mg galic acid L^− 1 respectively), thereby enhancing antioxidant capacity of the fermented beverages, yeast fermentations were characterized by a sharp decrease in sugars (1.43–2.19 mg 100 mL^− 1).Carotenoid compounds increased mainly in yeast monocultures (reaching up to approximately 150 mg L⁻¹), while remaining significantly lower in the other fermentations, with values ranging roughly between 12 and 30 mg L⁻¹. Volatile profiling identified 24 compounds, mainly dominated by acetic acid, ethyl octanoate and ethyl decanoate. The chemical composition of the beverages influenced the sensory profile, resulting in reduced sweetness and significant acidic notes, particularly in <i>L. plantarum</i> fermented juices. While this study successfully produced a novel functional goji-based fermented beverage, sensory analysis revealed room for improvement. This suggests that high potential for these innovative beverages to succeed through targeted product optimization, particularly in addressing the growing consumer interest in new functional fermented non-alcoholic drinks.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00217-026-05240-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Role of fermented and unfermented rooibos (Aspalathus linearis) in wet and dry marinades on oxidative status and sensory properties of pork meat subjected to different cooking methods","authors":"Marco Morosetti,&nbsp;Sara Barbarossa,&nbsp;Dalene De Beer,&nbsp;Benedetta Fanesi,&nbsp;Lama Ismaiel,&nbsp;Paolo Lucci,&nbsp;Luca Tiano,&nbsp;Andrea Frontini,&nbsp;Elisabetta Damiani,&nbsp;Patrick Orlando","doi":"10.1007/s00217-026-05279-2","DOIUrl":"10.1007/s00217-026-05279-2","url":null,"abstract":"<div><p>The antioxidant potential of <i>Aspalathus linearis</i> (rooibos), either fermented (FR) or unfermented (UR), used as a natural additive in dry or wet marinades for pork loin cooked using oven, pan, sous-vide and microwave methods was investigated. Their effect on lipid peroxidation (LP), on the content and oxidative status of coenzyme Q₁₀ (CoQ₁₀), on cooking-related meat weight loss and on sensorial properties were evaluated. All cooking methods increased LP endpoints, however pan-cooking lead to lower oxidation in terms of hydroperoxides (PV) (0.020 ± 0.003 mEq O₂/100 g in wet-marinated and 0.025 ± 0.000 mEq O₂/100 g in dry-marinated meat) and malondialdehyde (MDA) (0.124 ± 0.015 mg/100 g in wet-marinated and 0.163 ± 0.002 mg/100 g in dry-marinated meat). FR limited the increase in PV in dry-marinated samples cooked using oven (0.031 ± 0.007 mEq O₂/100 g), sous-vide (0.023 ± 0.007 mEq O₂/100 g) and microwave (0.260 ± 0.006 mEq O₂/100 g), compared to the control (C) (oven = 0.045 ± 0.004 mEq O₂/100 g sous-vide = 0.042 ± 0.003 mEq O₂/100 g, microwave = 0.046 ± 0.005 mEq O₂/100 g). For MDA levels, FR was effective in dry marinated and microwaved samples (C = 0.265 ± 0.061 mg/100 g, FR = 0.106 ± 0.026 mg/100 g) and in oven-cooked samples previously wet-marinated (C = 0.211 ± 0.022 mg/100 g, FR = 0.128 ± 0.005 mg/100 g) or dry-marinated (C = 0.247 ± 0.019 mg/100 g; FR = 0.159 ± 0.005 mg/100 g). Conversely, UR was able to significantly counteract both PV and MDA in oven, sous-vide and microwave cooked meat, regardless of the marinating technique used, proving even more effective in counteracting LP. On the other hand, FR seems to provide a more favourable sensory profile, particularly in samples cooked in the microwave with a wet marinade and in samples cooked in the oven with a dry marinade. The use of rooibos influenced CoQ₁₀ oxidative status under certain conditions. This study demonstrates the ability of rooibos extracts to modulate early oxidative changes occurring during meat cooking, indicating its potential use as a natural alternative to synthetic preservatives.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00217-026-05279-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Chemical mapping of Crocus sativus L. floral parts for assessing compositional consistency and auto-adulteration in saffron-based botanicals","authors":"Hilva Gjoni,&nbsp;Guillem Campmajó,&nbsp;Marco Biagi,&nbsp;Renato Bruni,&nbsp;Chiara Dall’Asta","doi":"10.1007/s00217-026-05263-w","DOIUrl":"10.1007/s00217-026-05263-w","url":null,"abstract":"<div><p>Phytochemical variability complicates quality assessment of botanical ingredients, particularly when authenticated reference materials are limited. Saffron (<i>Crocus sativus</i> L.), used both as a spice and as a food supplement ingredient, exhibits marked compositional differences across its floral parts, making evaluation of compositional balance essential for quality control and authentication. Here, saffron-based botanicals were profiled by LC-TWIMS-HRMS and evaluated within a floral part-anchored multivariate framework. Sixty-eight metabolites were tentatively identified across stigma, petal, and stamen reference tissues and nine commercial botanicals. Principal component analysis of log-transformed data revealed a dominant stigma-petal compositional gradient. Botanicals were projected onto the direction defined by anatomical centroids, and the robustness of directional estimates was assessed through compound-level and macro-class resampling. Orthogonal residual decomposition was used to characterize compositional variance not captured by the primary gradient. Most botanicals remained closely aligned with the stigma reference profile. A subset showed reproducible displacement toward petal-associated chemical patterns, driven by coordinated enrichment of flavonoids and anthocyanins relative to crocins, consistent with the presence of non-stigma floral material. These findings demonstrate that anatomically anchored chemical mapping combined with structured multivariate modelling can detect and characterize compositional deviations in saffron-based botanicals under conditions of limited reference availability.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00217-026-05263-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Revealing progressive wrinkling patterns in greenhouse cucumbers through portable Vis–NIR spectroscopy and unsupervised learning approaches","authors":"Nadia Saadati,&nbsp;Raziyeh Pourdarbani,&nbsp;Sajad Sabzi,&nbsp;Mohammad Vahedi Torshizi","doi":"10.1007/s00217-026-05276-5","DOIUrl":"10.1007/s00217-026-05276-5","url":null,"abstract":"<div><p>Cucumber fruits are highly susceptible to postharvest moisture loss and surface wrinkling, leading to rapid quality deterioration and reduced marketability. This study evaluated the potential of handheld visible–near infrared (Vis–NIR) spectroscopy combined with unsupervised machine learning algorithms for non-destructive identification of wrinkling stages during storage. A total of 220 greenhouse cucumbers were stored for 16 days under controlled conditions. Reflectance spectra were collected from the middle and stem-end regions in the 400–1100 nm range and analyzed in both the 400–1100 and 700–1100 nm intervals. Raw, Savitzky–Golay (SG), Multiplicative Scatter Correction (MSC), and Standard Normal Variate (SNV) preprocessing methods were evaluated. The clustering performances of Artificial Bee Colony (ABC), Fuzzy C-Means (FCM), Gaussian Mixture Models (GMM), K-Means, and Self-Organizing Maps (SOM) algorithms were assessed using the Davies–Bouldin (DB) and Silhouette indices. The results revealed significant effects of spectral preprocessing, sampling location, and wavelength range on clustering performance. Spectra collected from the stem-end region generally provided better class separability than those from the middle section. Among preprocessing methods, SG, MSC, and SNV improved clustering quality depending on the algorithm applied. K-Means, ABC, and SOM consistently outperformed FCM and GMM across most spectral scenarios. The highest clustering performance was achieved using K-Means combined with SG preprocessing in the stem-end region within the 700–1100 nm range, yielding a Silhouette coefficient of 0.844 and a Davies–Bouldin index of 0.472. Results further indicated that wrinkling progression followed a continuous physiological gradient, causing partial overlap among intermediate quality classes. Portable Vis–NIR spectroscopy coupled with unsupervised learning provides an effective tool for non-destructive monitoring of cucumber wrinkling and postharvest quality changes.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Understanding the challenges to reduce carbohydrates in foods: a review","authors":"Lívia Neves Santa Rosa,&nbsp;Renata Moraes Brito,&nbsp;Jucenir dos Santos Ferreira,&nbsp;Eliara Acipreste Hudson,&nbsp;Jaqueline de Paula Rezende,&nbsp;Márcia Cristina Teixeira Ribeiro Vidigal,&nbsp;Maria do Carmo Gouveia Peluzio,&nbsp;Ana Clarissa dos Santos Pires","doi":"10.1007/s00217-026-05267-6","DOIUrl":"10.1007/s00217-026-05267-6","url":null,"abstract":"<div><p>Growing interest in low-carbohydrate (low-carb) dietary patterns has stimulated research and industrial efforts toward reformulated foods with low carbohydrate content. However, carbohydrates are multifunctional components that contribute not only to sweetness, but also to structural stability, water distribution, viscosity, and sensory perception, making their reduction a major technological challenge. To address these challenges, this narrative review analyzed the main carbohydrate sources in processed foods, including sugars, starches, and wheat flour, highlighting the reformulation strategies, sensory implications, and technological barriers involved in developing low-carb alternatives. An exploratory keyword co-occurrence analysis was also conducted to contextualize current research trends and identify thematic gaps in literature. Current evidence suggests that no single ingredient or replacement strategy consistently reproduces the physicochemical and sensory properties of conventional carbohydrate-rich ingredients across different food systems. Instead, carbohydrate reduction frequently involves trade-offs between nutritional improvement, structural stability, processing behavior, and consumer acceptance. This review highlights the need for integrated and multidisciplinary approaches combining ingredient functionality, nutritional balance, processing strategies, and sensory optimization to support the development of technologically feasible and acceptable low-carb foods. This review provides a mechanistic and integrative framework to support future advances in low-carb food reformulation.</p></div>","PeriodicalId":549,"journal":{"name":"European Food Research and Technology","volume":"252 9","pages":""},"PeriodicalIF":4.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s00217-026-05267-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}