Monitoring complex media fermentations with near-infrared spectroscopy: comparison of different variable selection methods.

Near-infrared spectroscopy (NIRS) is known to be a suitable technique for rapid fermentation monitoring. Industrial fermentation media are complex, both chemically (ill-defined composition) and physically (multiphase sample matrix), which poses an additional challenge to the development of robust NIRS calibration models. We investigated the use of NIRS for at-line monitoring of the concentration of clavulanic acid during an industrial fermentation. An industrial strain of Streptomyces clavuligerus was cultivated at 200-L scale for the production of clavulanic acid. Partial least squares (PLS) regression was used to develop calibration models between spectral and analytical data. In this work, two different variable selection methods, genetic algorithms (GA) and PLS-bootstrap, were studied and compared with models built using all the spectral variables. Calibration models for clavulanic acid concentration performed well both on internal and external validation. The two variable selection methods improved the predictive ability of the models up to 20%, relative to the calibration model built using the whole spectra.

[1]  S. Vaidyanathan,et al.  Critical Evaluation of Models Developed for Monitoring an Industrial Submerged Bioprocess for Antibiotic Production Using Near‐Infrared Spectroscopy , 2000, Biotechnology progress.

[2]  R. Leardi,et al.  Variable selection for multivariate calibration using a genetic algorithm: prediction of additive concentrations in polymer films from Fourier transform-infrared spectral data , 2002 .

[3]  Vaidyanathan,et al.  At-line monitoring of a submerged filamentous bacterial cultivation using near-infrared spectroscopy. , 2000, Enzyme and microbial technology.

[4]  A. E. Bird,et al.  Spectrophotometric assay of clavulanic acid by reaction with imidazole , 1982 .

[5]  Linda M. Harvey,et al.  Deconvolution of near-infrared spectral information for monitoring mycelial biomass and other key analytes in a submerged fungal bioprocess , 2001 .

[6]  Hugo Kubinyi,et al.  Evolutionary variable selection in regression and PLS analyses , 1996 .

[7]  S. Engelsen,et al.  Interval Partial Least-Squares Regression (iPLS): A Comparative Chemometric Study with an Example from Near-Infrared Spectroscopy , 2000 .

[8]  Rainer Künnemeyer,et al.  Method of Wavelength Selection for Partial Least Squares , 1997 .

[9]  Jonas Johansson,et al.  Comparison of different variable selection methods conducted on NIR transmission measurements on intact tablets , 2003 .

[10]  S Vaidyanathan,et al.  Assessment of near-infrared spectral information for rapid monitoring of bioprocess quality. , 2001, Biotechnology and bioengineering.

[11]  Jean-Pierre Gauchi,et al.  Selecting both latent and explanatory variables in the PLS1 regression model , 2003 .

[12]  Hein Putter,et al.  The bootstrap: a tutorial , 2000 .

[13]  Paul Geladi,et al.  Interactive variable selection (IVS) for PLS. Part 1: Theory and algorithms , 1994 .

[14]  S. Arnold,et al.  Use of at‐line and in‐situ near‐infrared spectroscopy to monitor biomass in an industrial fed‐batch Escherichia coli process , 2002, Biotechnology and bioengineering.

[15]  D. Massart,et al.  Elimination of uninformative variables for multivariate calibration. , 1996, Analytical chemistry.

[16]  Riccardo Leardi,et al.  Application of genetic algorithm–PLS for feature selection in spectral data sets , 2000 .

[17]  L. Buydens,et al.  Development of robust calibration models in near infra-red spectrometric applications , 2000 .