Advertisement

Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Modeling of the pyruvate production with Escherichia coli: comparison of mechanistic and neural networks-based models

Abstract

Three different models: the unstructured mechanistic black-box model, the input–output neural network-based model and the externally recurrent neural network model were used to describe the pyruvate production process from glucose and acetate using the genetically modified Escherichia coli YYC202 ldhA::Kan strain. The experimental data were used from the recently described batch and fed-batch experiments [ Zelić B, Study of the process development for Escherichia coli-based pyruvate production. PhD Thesis, University of Zagreb, Faculty of Chemical Engineering and Technology, Zagreb, Croatia, July 2003. (In English); Zelić et al. Bioproc Biosyst Eng 26:249–258 (2004); Zelić et al. Eng Life Sci 3:299–305 (2003); Zelić et al Biotechnol Bioeng 85:638–646 (2004)]. The neural networks were built out of the experimental data obtained in the fed-batch pyruvate production experiments with the constant glucose feed rate. The model validation was performed using the experimental results obtained from the batch and fed-batch pyruvate production experiments with the constant acetate feed rate. Dynamics of the substrate and product concentration changes was estimated using two neural network-based models for biomass and pyruvate. It was shown that neural networks could be used for the modeling of complex microbial fermentation processes, even in conditions in which mechanistic unstructured models cannot be applied.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Abbreviations

c A :

acetate concentration (g L−1)

c A,0 :

acetate concentration in the feed (g L−1)

c G :

gucose concentration (g L−1)

c G,0 :

glucose concentration in the feed (g L−1)

c P :

pyruvate concentration (g L−1)

c P,MAX :

critical pyruvate concentration above which reaction cannot proceed (g L−1)

c X :

biomass concentration (g L−1)

K P :

inhibition constant of Jerusalimsky (g L−1)

K S A :

monod growth constant for acetate (g L−1)

K S G :

monod growth constant for glucose (g L−1)

m A :

Maintenance coefficient for growth on acetate (g g−1 h−1)

m G :

maintenance coefficient for growth on glucose (g g−1 h−1)

q V :

volumetric flow rate (L h−1)

q VA :

volumetric flow rate of acetate (L h−1)

q VG :

volumetric flow rate of glucose (L h−1)

r A :

specific rate of acetate consumption (g g−1 h−1)

r G :

specific rate of glucose consumption (g g−1 h−1)

r P :

specific rate of pyruvate production (g g−1 h−1)

t :

time (h)

V :

reaction (broth) volume (L)

Y P/G :

yield coefficient pyruvate from glucose (g g−1)

Y X/A :

yield coefficient biomass from acetate (g g−1)

Y X/A,MAX :

maximum yield coefficient biomass from acetate (g g−1)

Y X/G :

yield coefficient biomass from glucose (g g−1)

Y X/G,MAX :

maximum yield coefficient biomass from glucose (g g−1)

u :

input variables (–)

x :

current output state (–)

μ:

specific growth rate (h−1)

μMAX :

maximum specific growth rate (h−1)

References

  1. 1.

    Zelić B (2003) Study of the process development for Escherichia coli based pyruvate production. PhD Thesis, University of Zagreb, Faculty of Chemical Engineering and Technology, Zagreb, Croatia (In English)

  2. 2.

    Zelić B, Vasić-Rački Đ, Wandrey C, Takors R (2004) Modeling of the pyruvate production with Escherichia coli in a fed-batch bioreactor. Bioproc Biosyst Eng 26:249–258

  3. 3.

    Prasad V, Bequette BW (2003) Nonlinear system identification and model reduction using artificial neural networks. Comp Chem Eng 27:1741–1754

  4. 4.

    DiMassimo C, Montague GA, Willis MJ, Morris AJ, Tham MT (1992) Enhancing industrial bioprocess monitoring through artificial neural networks. IFAC Modeling and Control of Biotech Proc, Keystone, Colorado, 395–398

  5. 5.

    Linko P, Zhu YH (1991) Neural network programming in bioprocess variable estimation and state prediction. J Biotechnol 21:253–270

  6. 6.

    James S, Legge R, Budmann H (2002) Comparative study of black-box and hybrid estimation methods in fed-batch fermentation. J Proc Control 12:113–121

  7. 7.

    Linko S, Luopa J, Zhu YH (1997) Neural networks as “software sensors” in enzyme production. J Biotechnol 52:257–266

  8. 8.

    Martin GD (1997) Consider soft sensors. Chem Eng Prog 66–70

  9. 9.

    Dochain D, Perrier M (1997) Dynamical modeling, analysis, monitoring and control design for nonlinear bioprocesses. Adv Biochem Eng 56:149–197

  10. 10.

    Kapil GG, Mehra S, Gomes J (2005) On-line adaptation of neural network for bioprocess control. Comp Chem Eng 29:1047–1057

  11. 11.

    Baughman DR, Liu YA (1995) Neural networks in bioprocessing and chemical engineering. Academic, San Diego, CA

  12. 12.

    Masters T (1995) Neural, novel & hybrid algorithms for continually running fully recurrent neural networks. Wiley, New York

  13. 13.

    Gosak D, Vampola M (2001) Generic neural model control of batch distillation process. IchemE symposium “Advances in Process Control”, CD-ROM

  14. 14.

    Elman JL (1990) Finding structure in time. Cogn Sci 14:179–211

  15. 15.

    Jordan MI (1986) Attractor dynamics and parallelism in a connectionist sequential machine. The eight annual conference of the cognitive science society, Amherst, USA, 531–546

  16. 16.

    Perretto P, Niez JJ (1986) Stochastic dynamics of neural networks. IEEE Trans Syst Man Cybern 16:73–83

  17. 17.

    Nerrand O (1994) Training recurrent neural networks: why and how? An illustration in dynamical process modelling. IEEE Trans Neural Netw 5:178–184

  18. 18.

    Warnes MR, Glassey J, Montague GA, Kara B (1998) Application of radial basis function and feedforward artificial neural networks to the Escherichia coli fermentation process. Neurocomputing 20:67–82

  19. 19.

    Bachinger TH, Martensson P, Mandenius CF (1998) Estimation of biomass and specific growth rate in a recombinant Escherichia coli batch cultivation process using a chemical multisensor array. J Biotechnol 60:55–66

  20. 20.

    Hajmeer MN, Basheer IA (2003) A hybrid Bayesian–neural network approach for probabilistic modeling of bacterial growth/no-growth interface. Int J Food Microbiol 82:233–243

  21. 21.

    Lee J, Lee SY, Park S, Middelberg APJ (1999) Control of fed-batch fermentations. Biotechnol Adv 17:29–48

  22. 22.

    Vlassides S, Ferrier JG, Block DE (2001) Using historical data for bioprocess optimization: modeling wine characteristics using artificial neural networks and archived process information. Biotechnol Bioeng 73:55–68

  23. 23.

    Zelić B, Gerharz T, Bott M, Vasić-Rački Đ, Wandrey C, Takors R (2003) Fed-batch process for pyruvate production by recombinant Escherichia coli YYC202 strain. Eng Life Sci 3:299–305

  24. 24.

    Zelić B, Gostović S, Vuorilehto K, Vasić-Rački Đ, Takors R (2004) Process strategies to enhance pyruvate production with recombinant Escherichia coli: from repetitive fed-batch to in situ product recovery with fully integrated electrodialysis. Biotechnol Bioeng 85:638–646

  25. 25.

    Roels JA (1983) Energetics and kinetics in biotechnology. Elsevier, Amsterdam

  26. 26.

    Sander A, Skansi D, Bolf N (2003) Heat and mass transfer models in convection drying of clay slabs, Ceramic Int 29:641–653

  27. 27.

    De Boor C (1978) A practical guide to splines. Springer Verlag, Berlin Heidelberg New York

  28. 28.

    Novakovic B, Majetic D, Siroki M (1998) Artifical neural networks. University Press, Zagreb

  29. 29.

    Neuralware, Neural computing—A technology handbook for professional II/plus and neuralworks explorer, NeuralWare Inc., Pittsburgh, USA, 1996

  30. 30.

    Patnaik PR (1997) A recurrent neural network for a fed-batch fermentation with recombinant Escherichia coli subject to inflow disturbances. Proc Biochem 32:391–400

  31. 31.

    Bishop CM (1995) Neural network for pattern recognition. Oxford University Press, New York

Download references

Acknowledgments

All experiments were performed at the Institute of Biotechnology 2, Research Center Jülich, Germany. The authors are indebted to Prof. Dr Christian Wandrey and Dr Ralf Takors for generous support of this work. This work was partially funded by the Croatian Ministry of Science, Education and Sport, contract grant number 0125 021.

Author information

Correspondence to B. Zelić.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Zelić, B., Bolf, N. & Vasić-Rački, Đ. Modeling of the pyruvate production with Escherichia coli: comparison of mechanistic and neural networks-based models. Bioprocess Biosyst Eng 29, 39–47 (2006). https://doi.org/10.1007/s00449-006-0054-3

Download citation

Keywords

  • Pyruvate
  • Escherichia coli
  • Unstructured “black-box” model
  • Input–output neural network-based model
  • Externally recurrent neural network model