Chemometric models for on-line process monitoring have become well established in pharmaceutical bioprocesses. The main drawback is the required calibration effort and the inflexibility regarding system or process changes. So, a recalibration is necessary whenever the process or the setup changes even slightly. With a large and diverse Raman dataset, however, it was possible to generate generic partial least squares regression models to reliably predict the concentrations of important metabolic compounds, such as glucose-, lactate-, and glutamine-indifferent CHO cell cultivations. The data for calibration were collected from various cell cultures from different sites in different companies using different Raman spectrophotometers. In testing, the developed “generic” models were capable of predicting the concentrations of said compounds from a dilution series in FMX-8 mod medium, as well as from an independent CHO cell culture. These spectra were taken with a completely different setup and with different Raman spectrometers, demonstrating the model flexibility. The prediction errors for the tests were mostly in an acceptable range (<10% relative error). This demonstrates that, under the right circumstances and by choosing the calibration data carefully, it is possible to create generic and reliable chemometric models that are transferrable from one process to another without recalibration.

There is increasing demand for cereal-based probiotic fermented beverages as an alternative to dairy-based products due to their limitations. However, analyzing and monitoring the fermentation process is usually time consuming, costly, and labor intensive. This research therefore aims to apply two-dimensional (2D)-fluorescence spectroscopy coupled with partial least-squares regression (PLSR) and artificial neural networks (ANN) for the on-line quantitative analysis of cell growth and concentrations of lactic acid and glucose during the fermentation of a teff-based substrate. This substrate was inoculated with mixed strains of Lactiplantibacillus plantarum A6 (LPA6) and Lacticaseibacillus rhamnosus GG (LCGG). The fermentation was performed under two different conditions: condition 1 (7 g/100 mL substrate inoculated with 6 log cfu/mL) and condition 2 (4 g/100 mL substrate inoculated with 6 log cfu/mL). For the prediction of LPA6 and LCGG cell growth, the relative root mean square error of prediction (pRMSEP) was measured between 2.5 and 4.5%. The highest pRMSEP (4.5%) was observed for the prediction of LPA6 cell growth under condition 2 using ANN, but the lowest pRMSEP (2.5%) was observed for the prediction of LCGG cell growth under condition 1 with ANN. A slightly more accurate prediction was found with ANN under condition 1. However, under condition 2, a superior prediction was observed with PLSR as compared to ANN. Moreover, for the prediction of lactic acid concentration, the observed values of pRMSEP were 7.6 and 7.7% using PLSR and ANN, respectively. The highest error rates of 13 and 14% were observed for the prediction of glucose concentration using PLSR and ANN, respectively. Most of the predicted values had a coefficient of determination (R²) of more than 0.85. In conclusion, a 2D-fluorescence spectroscopy combined with PLSR and ANN can be used to accurately monitor LPA6 and LCGG cell counts and lactic acid concentration in the fermentation process of a teff-based substrate. The prediction of glucose concentration, however, showed a rather high error rate.

The demand for probiotic bacteria-fermented food products is increasing; however, the monitoring of the fermentation process is still challenging when using conventional approaches. A classical approach requires a large amount of offline data to calibrate a chemometric model using fluorescence spectra. Fluorescence spectra provide a wide range of online information during the process of cultivation, but they require a large amount of offline data (which involves laborious work) for the calibration procedure when using a classical approach. In this study, an alternative model-based calibration approach was used to predict biomass (the growth of Lactiplantibacillus plantarum A6 (LPA6) and Lacticaseibacillus rhamnosus GG (LCGG)), glucose, and lactic acid during the fermentation process of a teff-based substrate inoculated with mixed strains of LPA6 and LCGG. A classical approach was also applied and compared to the model-based calibration approach. In the model-based calibration approach, two-dimensional (2D) fluorescence spectra and offline substituted simulated data were used to generate a chemometric model. The optimum microbial specific growth rate and chemometric model parameters were obtained simultaneously using a particle swarm optimization algorithm. The prediction errors for biomass, glucose, and lactic acid concentrations were measured between 6.1 and 10.5%; the minimum error value was related to the prediction of biomass and the maximum one was related to the prediction of glucose using the model-based calibration approach. The model-based calibration approach and the classical approach showed similar results. In conclusion, the findings showed that a model-based calibration approach could be used to monitor the process state variables (i.e., biomass, glucose, and lactic acid) online in the fermentation process of a teff-based substrate inoculated with mixed strains of LPA6 and LCGG. However, glucose prediction showed a high error value.

Background and Objectives: Research in crop science is usually conducted in studies with a small number of plants and crops. The aim of this project was to investigate whether near‐infrared spectroscopy (NIRS) can contribute to the analysis of such small data sets. For this purpose, NIR spectra of a small sample set of whole‐grain wheat flours of two winter wheat cultivars harvested from greenhouse and field experiments with different fertilization management were analyzed. Findings: Principal component analysis (PCA) facilitated the discrimination of samples based on fertilization, baking volume, and growing conditions for each wheat cultivar. In addition, PCA allowed the differentiation of wheat cultivars in a data set containing both cultivars. Spectra–structure relationships were also established. Accurate calibrations with R ² of .98, .98, and .54 were obtained for fertilization, crude protein concentration, and baking volume, respectively, even for a small data set of 10 samples of wheat cultivar Discus. Conclusion: The NIR spectra of whole‐grain wheat flour samples provided high‐resolution information on various parameters such as fertilization, baking volume, wheat cultivar, and growing site. This study demonstrates that NIRS is a valuable and efficient tool for the analysis of small data sets in crop science. This illustrates the effectiveness of the elegant NIRS method. Significance and Novelty: In this study, small data sets of whole‐grain wheat flour from crop science experiments were analyzed by NIRS. Characteristic changes of the NIR spectra based on applied fertilizer amount, crude protein concentration, baking volume, growing site, and wheat cultivar were revealed. The detection of cause‐and‐effect relationships between spectra and chemical properties of samples contributes to a better understanding of NIR spectra of wheat flours in an interpretative way.