Measurement of fermentation parameters of Chinese rice wine using Raman spectroscopy combined with linear and non-linear regression methods

In-line monitoring of bioreactor by Raman spectroscopy: Direct use of a standard-based model through cell-scattering correction

Raman spectroscopy and machine learning have become popular in in-line monitoring of bioreactors. However, traditional modeling processes typically entail extensive fermentation batches to collect learning datasets, which are significantly time-consuming and laborious. In addition, these models are limited to configurations with the same conditions as the training batches. The present work proposes a reproducible and adaptable modeling approach by combining standard spectra as a training dataset, with a simple means of correcting for cell scattering. Alcoholic fermentation by Saccharomyces cerevisiae is used as a benchmark. Initially, a partial least squares (PLS) regression model was developed based on the spectra of pure solutions of glucose and ethanol. Then, a mathematical expression was defined to estimate yeast concentration, allowing the correction of Raman intensity attenuated by cell scattering. The corrected spectra demonstrate close alignment with reference spectra in both shape and intensity. Validation of the methodology was conducted across numerous batches and one fed-batch bioreactor. As a result, the developed method enables the simultaneous monitoring of glucose, ethanol, and yeast concentrations, effectively addressing the challenge of implementing an independent standards based PLS model to manage the intricate compositional dynamics in bio-processes. The conclusion underscores the effectiveness of the proposed method and offers new prospects in biotechnological industries.

Spectral encoder to extract the efficient features of Raman spectra for reliable and precise quantitative analysis

Raman spectroscopy has become a powerful analytical tool highly demanded in many applications such as microorganism sample analysis, food quality control, environmental science, and pharmaceutical analysis, owing to its non-invasiveness, simplicity, rapidity and ease of use. Among them, quantitative research using Raman spectroscopy is a crucial application field of spectral analysis. However, the entire process of quantitative modeling largely relies on the extraction of effective spectral features, particularly for measurements on complex samples or in environments with poor spectral signal quality. In this paper, we propose a method of utilizing a spectral encoder to extract effective spectral features, which can significantly enhance the reliability and precision of quantitative analysis. We built a latent encoded feature regression model; in the process of utilizing the autoencoder for reconstructing the spectrometer output, the latent feature obtained from the intermediate bottleneck layer is extracted. Then, these latent features are fed into a deep regression model for component concentration prediction. Through detailed ablation and comparative experiments, our proposed model demonstrates superior performance to common methods on single-component and multi-component mixture datasets, remarkably improving regression precision while without needing user-selected parameters and eliminating the interference of irrelevant and redundant information. Furthermore, in-depth analysis reveals that latent encoded feature possesses strong nonlinear feature representation capabilities, low computational costs, wide adaptability, and robustness against noise interference. This highlights its effectiveness in spectral regression tasks and indicates its potential in other application fields. Sufficient experimental results show that our proposed method provides a novel and effective feature extraction approach for spectral analysis, which is simple, suitable for various methods, and can meet the measurement needs of different real-world scenarios.

Raman spectroscopy applied to online monitoring of a bioreactor: Tackling the limit of detection

An in-situ monitoring model of alcoholic fermentation based on Raman spectroscopy was developed in this study. The optimized acquisition parameters were an 80 s exposure time with three accumulations. Standard solutions were prepared and used to populate a learning database. Two groups of mixed solutions were prepared for a validation database to simulate fermentation at different conditions. First, all spectra of the standards were evaluated by principal component analysis (PCA) to identify the spectral features of the target substances and observe their distribution and outliers. Second, three multivariate calibration models for prediction were developed using the partial least squares (PLS) method, either on the whole learning database or subsets. The limit of detection (LOD) of each model was estimated by using the root mean square error of cross validation (RMSECV), and the prediction ability was further tested with both validation datasets. As a result, improved LODs were obtained: 0.42 and 1.55 g·L-1 for ethanol and glucose using a sub-learning dataset with a concentration range of 0.5 to 10 g·L-1. An interesting prediction result was obtained from a cross-mixed validation set, which had a root mean square error of prediction (RMSEP) for ethanol and glucose of only 3.21 and 1.69, even with large differences in mixture concentrations. This result not only indicates that a model based on standard solutions can predict the concentration of a mixed solution in a complex matrix but also offers good prospects for applying the model in real bioreactors.

Non-destructive prediction of colour and water-related properties of frozen/thawed beef meat by Raman spectroscopy coupled multivariate calibration

Freeze-thaw accelerated the colour deterioration of beef with the increase of colour b* and the decrease of colour a* values (P < 0.05). The maximum exudate loss reached 22 % after the seventh freeze-thaw. A strong correlation between the transversal relaxation time T₂₁ and thawing loss may mean that T₂₁ water contributed to the exudate loss during freeze-thaw. Afterwards, competitive adaptive reweighted sampling-partial least square (CARS-PLS) has the best prediction in thawing loss of frozen/thawed beef with correlation coefficients of prediction (R_p) of 0.971, and root mean square error of prediction (RMSEP) of 1.436. Besides, Uninformative variable elimination-partial least squares (UVE-PLS) showed good prediction effects on colour values (R_p = 0.932 - 0.994) and water content (R_p = 0.928, RMSEP = 0.582) of frozen/thawed beef. Therefore, this work demonstrated that Raman spectroscopy coupled with multivariate calibration has a good ability for non-destructive prediction in colour and water-related properties of frozen/thawed beef.

Identification of Chinese red wine origins based on Raman spectroscopy and deep learning

In this study, we combined Raman spectroscopy with deep learning for the first time to establish an accurate, simple, and fast method to identify the origin of red wines. We collected Raman spectra from 200 red wine samples of the Cabernet Sauvignon variety from four different origins with a portable Raman spectrometer. The red wine samples, made in 2021, were from the same producer in China. Differences were found by analyzing the Raman spectra of red wine samples. These differences are mainly caused by ethanol, carboxylic acids, and polyphenols. After further analysis, for different origins, the different performances of these substances on the Raman spectrum are related to the climate and geographical conditions of the origin. The Raman spectra were analyzed by principal component analysis (PCA). The data with PCA dimensionality reduction were imported into an artificial neural network (ANN), multifeature fusion convolutional neural network (MCNN), GoogLeNet, and residual neural network (ResNet) to establish red wine origin identification models. The classification results of the model prove that climate, geography, and other conditions can provide support for the classification of red wine origin. The experiments showed that all four models performed well, among which MCNN performed the best with 93.2% classification accuracy, and the area under the curve (AUC) was 0.987. This study provides a new means to classify the origin of red wine and opens up new ideas for identifying origins in the food field.

In situ Raman quantitative monitoring of methanogenesis: Culture experiments of a deep-sea cold seep methanogenic archaeon

Gas production from several metabolic pathways is a necessary process that accompanies the growth and central metabolism of some microorganisms. However, accurate and rapid nondestructive detection of gas production is still challenging. To this end, gas chromatography (GC) is primarily used, which requires sampling and sample preparation. Furthermore, GC is expensive and difficult to operate. Several researchers working on microbial gases are looking forward to a new method to accurately capture the gas trends within a closed system in real-time. In this study, we developed a precise quantitative analysis for headspace gas in Hungate tubes using Raman spectroscopy. This method requires only a controlled focus on the gas portion inside Hungate tubes, enabling nondestructive, real-time, continuous monitoring without the need for sampling. The peak area ratio was selected to establish a calibration curve with nine different CH4–N2 gaseous mixtures and a linear relationship was observed between the peak area ratio of methane to nitrogen and their molar ratios (A(CH4)/A(N2) = 6.0739 × n(CH4)/n(N2)). The results of in situ quantitative analysis using Raman spectroscopy showed good agreement with those of GC in the continuous monitoring of culture experiments of a deep-sea cold seep methanogenic archaeon. This method significantly improves the detection efficiency and shows great potential for in situ quantitative gas detection in microbiology. It can be a powerful complementary tool to GC.

Olfactory visualization sensor system based on colorimetric sensor array and chemometric methods for high precision assessing beef freshness

Beef is easily spoiled, resulting in foodborne illness and high societal costs. This study proposed a novel olfactory visualization system based on colorimetric sensor array and chemometric methods to detect beef freshness. First, twelve color-sensitive materials were immobilized on a hydrophobic platform to acquire scent information of beef samples according to solvatochromic effects. Second, machine vision algorithms were used to extract the scent fingerprints, and principal component analysis (PCA) was employed to compress the feature dimensions of the fingerprints. Finally, four qualitative models, k-nearest neighbor, extreme learning machine, support vector machine (SVM), and random forest, were constructed to evaluate the beef freshness according to the value of total volatile basic nitrogen (TVB-N) and total viable counts (TVC). Results demonstrated that SVM had a preferable prediction ability, with 95.83% and 95.00% precision in the training and prediction sets, respectively. The results revealed that the simple constructed olfactory visualization sensor system could rapidly, robustly, and accurately assess beef freshness.

Flavor compounds with high odor activity values (OAV > 1) dominate the aroma of aged Chinese rice wine (Huangjiu) by molecular association

Aging is an essential operation to perfect the flavor quality of Hungjiu. In this study, formation mechanism of flavor compounds responsible for the characteristic flavor of aged Huangjiu was investigated. The contents of umami and bitter free amino acids (FAA) increased with the storage period prolonged, while that of sweet FAA showed downward trend. Gas chromatograph-mass spectrometry and principal component analysis indicated that the volatile flavor compounds with OAV exceed 1, especially middle-chain fatty-acid-ethyl-esters and aromatic compounds, dominated the characteristic flavor of aged Huangjiu. Low field-NMR was firstly applied to characterize the molecular association between water and dissolved flavor compounds in aged Huangjiu. The results showed that basic amino acids contributed greatly to the flavor formation of aged Huangjiu via molecular association. In addition, the molecular association significantly promoted the accumulation of flavor compounds with OAV > 1, especially ethyl esters.

Histamine Causes Pyroptosis of Liver by Regulating Gut-Liver Axis in Mice

Huangjiu usually caused rapid-drunkenness and components such as β-benzyl ethanol (β-be), isopentanol (Iso), histamine (His), and phenethylamine (PEA) have been reported linked with intoxication. However, the destructive effect of these components on gut microbiota and liver is unclear. In this study, we found oral treatment of these components, especially His, stimulated the level of oxidative stress and inflammatory cytokines in liver and serum of mice. The gut microbiota community was changed and the level of lipopolysaccharide (LPS) increased significantly. Additionally, cellular pyroptosis pathway has been assessed and correlation analysis revealed a possible relationship between gut microbiota and liver pyroptosis. We speculated oral His treatment caused the reprogramming of gut microbiota metabolism, and increased LPS modulated the gut-liver interaction, resulting in liver pyroptosis, which might cause health risks. This study provided a theoretical basis for the effect of Huangjiu, facilitating the development of therapeutic and preventive strategies for related inflammatory disorders.

Quantitative TLC-SERS detection of histamine in seafood with support vector machine analysis

Scombroid fish poisoning caused by histamine intoxication is one of the most prevalent allergies associated with seafood consumption in the United States. Typical symptoms range from mild itching up to fatal cardiovascular collapse seen in anaphylaxis. In this paper, we demonstrate rapid, sensitive, and quantitative detection of histamine in both artificially spoiled tuna solution and real spoiled tuna samples using thin layer chromatography in tandem with surface-enhanced Raman scattering (TLC-SERS) sensing methods, enabled by machine learning analysis based on support vector regression (SVR) after feature extraction with principal component analysis (PCA). The TLC plates used herein, which were made from commercial food-grade diatomaceous earth, served simultaneously as the stationary phase to separate histamine from the blended tuna meat and as ultra-sensitive SERS substrates to enhance the detection limit. Using a simple drop cast method to dispense gold colloidal nanoparticles onto the diatomaceous earth plate, we were able to directly detect histamine concentration in artificially spoiled tuna solution down to 10 ppm. Based on the TLC-SERS spectral data of real tuna samples spoiled at room temperature for 0 to 48 hours, we used the PCA-SVR quantitative model to achieve superior predictive performance exceling traditional partial least squares regression (PLSR) method. This work proves that diatomaceous earth based TLC-SERS technique combined with machine-learning analysis is a cost-effective, reliable, and accurate approach for on-site detection and quantification of seafood allergen to enhance food safety.

Rapid determination by near infrared spectroscopy of theaflavins-to-thearubigins ratio during Congou black tea fermentation process

The theaflavin-to-thearubigin ratio (TF/TR) is an important parameter for evaluating the degree of fermentation and quality characteristics of Congou black tea. Near infrared (NIR) spectroscopy, one of the most promising techniques for evaluating large-scale tea processing quality, in association with chemometrics, can be used as a selection tool when a fast determination of the requested parameters is required. The aim of this work is to develop a unique model for the determination of TF/TR. First, 11 key wavelength variables were screened by synergy interval partial least-squares regression (SI-PLS) and competitive adaptive reweighted sampling (CARS). Based on these characteristic variables, a new extreme learning machine (ELM) combined with an adaptive boosting (ADABOOST) algorithm (ELM-ADABOOST) was applied to construct the nonlinear prediction model for TF/TR, and an independent external set was used for the validation. A determinate coefficient (R_p²) of 0.893, root mean square error of prediction (RMSEP) of 0.0044, RSD below 10%, and RPD above 3 were acquired in the prediction model. These results demonstrate that NIR can be used to rapidly determine the TF/TR value during fermentation, and it effectively simplify the model and improve the prediction accuracy when combined with the SI-CARS variable.

Detection of ethyl carbamate in liquors using surface-enhanced Raman spectroscopy

Ethyl carbamate (EC), a potential carcinogen, can be formed during the fermentation and storage of alcoholic beverages. In this work, quantitative detection of EC in alcoholic beverages by using surface-enhanced Raman spectroscopy (SERS) is reported. Flower-shaped silver nanostructure substrates and silver nanocube substrates were prepared and employed as SERS platform. Flower-like silver substrates had better Raman enhancement effect on EC and were selected for further EC detection. In EC SERS spectra based on flower-shaped silver substrates, the strongest and reproducible characteristic band at 857 cm^-1 was chosen for establishing a linear regression model in the concentrations ranging from 10^-5 to 10^-9 M, which effectively extended the application scope of the quantitative model for determination EC. Furthermore, a real alcoholic beverage was tested to verify the feasibility and reliability of the method.

Discrimination of wine lactic acid bacteria by Raman spectroscopy

Species of Lactobacillus, Pediococcus, Oenococcus, and Leuconostoc play an important role in winemaking, as either inoculants or contaminants. The metabolic products of these lactic acid bacteria have considerable effects on the flavor, aroma, and texture of a wine. However, analysis of a wine's microflora, especially the bacteria, is rarely done unless spoilage becomes evident, and identification at the species or strain level is uncommon as the methods required are technically difficult and expensive. In this work, we used Raman spectral fingerprints to discriminate 19 strains of Lactobacillus, Pediococcus, and Oenococcus. Species of Lactobacillus and Pediococcus and strains of O. oeni and P. damnosus were classified with high sensitivity: 86-90 and 84-85%, respectively. Our results demonstrate that a simple, inexpensive method utilizing Raman spectroscopy can be used to accurately identify lactic acid bacteria isolated from wine.