Bioprocess in‐line monitoring using Raman spectroscopy and Indirect Hard Modeling (IHM): A simple calibration yields a robust model

Abstract

To increase the process productivity and product quality of bioprocesses, the in‐line monitoring of critical process parameters is highly important. For monitoring substrate, metabolite, and product concentrations, Raman spectroscopy is a commonly used Process Analytical Technology (PAT) tool that can be applied in‐situ and non‐invasively. However, evaluating bioprocess Raman spectra with a robust state‐of‐the‐art statistical model requires effortful model calibration. In the present study, we in‐line monitored a glucose to ethanol fermentation by Saccharomyces cerevisiae (S. cerevisiae) using Raman spectroscopy in combination with the physics‐based Indirect Hard Modeling (IHM) and showed successfully that IHM is an alternative to statistical models with significantly lower calibration effort. The IHM prediction model was developed and calibrated with only 16 Raman spectra in total, which did not include any process spectra. Nevertheless, IHM's root mean square errors of prediction (RMSEPs) for glucose (3.68 g/L) and ethanol (1.69 g/L) were comparable to the prediction quality of similar studies that used statistical models calibrated with several calibration batches. Despite our simple calibration, we succeeded in developing a robust model for evaluating bioprocess Raman spectra.

The Good pH probe: non-invasive pH in-line monitoring using Good buffers and Raman spectroscopy

“…The prediction quality of the Good pH probe was not degraded by the additional spectral superimposing component, since it uses the physics-based IHM for spectral evaluation. A calibrated IHM prediction model can be transferred to a more complex system by extending the model with the respective pure component models [ 34 ]. This transfer does not affect the recently determined calibration factors as long as potential newly occurring nonlinear mixture effects are covered by additional adjustable peak parameters in the model.…”

“…What remains to be shown is whether the Good pH probe can be successfully applied to more complex bioprocesses beyond enzyme reactions, such as E. coli fermentations. In this regard, our previous study has already demonstrated that the combination of Raman spectroscopy and IHM is capable of monitoring other critical process parameters in glucose to ethanol fermentations by Saccharomyces cerevisiae ( S. cerevisiae ) [ 34 ].…”

“…To be more precise, we evaluated the Good pH probe’s calibration with the RMSE of cross-validation (RMSECV). The RMSECV was calculated using a leave-one-out cross-validation approach, in which we determined the for each data point in the calibration dataset, while we used the remaining dataset for the calibration of the Good pH probe [ 34 , 36 ]. To evaluate the characterization and application of the Good pH probe, we determined the RMSE of prediction (RMSEP).…”

“…For this purpose, we combined MOPS as pH indicator with Raman spectroscopy as spectroscopic technique and with Indirect Hard Modeling (IHM) as spectral evaluation model. Since IHM is physics-based, it can provide robust models for evaluating bioprocess Raman spectra with significantly reduced calibration effort compared with state-of-the-art statistical models [ 33 , 34 ].…”

“…To the best of our knowledge, this is the first application of a Good buffer as pH indicator, and the first use of IHM to evaluate Raman spectra for spectroscopic pH monitoring. With respect to our previous study, in which we in-line monitored the substrate and product mass fractions in a bioprocess by combining Raman spectroscopy with IHM, the Good pH probe adds the pH value to the list of critical process parameters that can be monitored using this combination [ 34 ]. We thoroughly characterized the Good pH probe in regard to its reversibility, temperature dependence, sensitivity to ionic strength, and applicability in systems with components whose Raman spectra strongly superimpose the spectral features of MOPS.…”