Monitoring of complex industrial bioprocesses for metabolite concentrations using modern spectroscopies and machine learning: application to gibberellic acid production.

Two rapid vibrational spectroscopic approaches (diffuse reflectance-absorbance Fourier transform IR [FT-IR] and dispersive Raman spectroscopy), and one mass spectrometric method based on in vacuo Curie-point pyrolysis (PyMS), were investigated in this study. A diverse range of unprocessed, industrial fed-batch fermn. broths contg. the fungus Gibberella fujikuroi producing the natural product gibberellic acid, were analyzed directly without a priori chromatog. sepn. Partial least squares regression (PLSR) and artificial neural networks (ANNs) were applied to all of the information-rich spectra obtained by each of the methods to obtain quant. information on the gibberellic acid titer. These ests. were of good precision, and the typical root-meansquare error for predictions of concns. in an independent test set was < 10% over a very wide titer range from 0 to 4925 ppm. However, although PLSR and ANNs are very powerful techniques they are often described as \"black box\" methods because the information they use to construct the calibration model is largely inaccessible. Therefore, a variety of novel evolutionary computation-based methods, including genetic algorithms and genetic programming, were used to produce models that allowed the detn. of those input variables that contributed most to the models formed, and to observe that these models were predominantly based on the concn. of gibberellic acid itself. This is the first time that these three modern anal. spectroscopies, in combination with advanced chemometric data anal., have been compared for their ability to analyze a real com. bioprocess. The results demonstrate unequivocally that all methods provide very rapid and accurate ests. of the progress of industrial fermns., and indicate that, of the three methods studied, Raman spectroscopy is the ideal bioprocess monitoring method because it can be adapted for online anal.