Çok Katmanlı Perceptron Yapay Sinir Ağı Kullanılarak Maya Hücrelerinin Yaşam Döngüsü Parametrelerinin Araştırılması
Gıda endüstrisinde maya hücrelerinin büyüme parametrelerinin incelenmesi hem zaman hem de işçilik maliyetlerinde artışa neden olmaktadır. Bu maliyetleri azaltmak için simülasyon modelleri öne sürülebilir. Bu çalışmada Çok Katmanlı Perceptron Sinir Ağı (MLPNN) kullanarak Saccharomyces cerevisiae'nin büyüme döngüsü parametreleri için bir simülasyon modeli tasarlanması amaçlanmıştır. Bu modelde, ekim zamanı girdi parametresi olarak tanımlanmışken, saatteki büyüme oranı ve hücre sayısı çıktı parametreleri olarak belirlenmiştir. Tasarlanan modelde iki gizli katmanlı geri yayılımlı sinir ağı tercih edilmiştir. İlk gizli katmanda 10 düğümün kullanılırken ikinci gizli katmanı 2 düğüm kullanılmıştır. Modelin eğitimi için 144 deneysel veri kullanılırken, bu deneysel verilerin 72'si eğitilmiş modeli test etmek için kullanılmıştır. Geliştirilen model, büyüme eğrisi ve büyüme hızı için hem eğitim hem de test aşamasında yüksek bir korelasyon göstermiştir (büyüme eğrisi için, R2 training = 0,9993 ve R2 test = 0,9993; büyüme oranı için R2 training = 0.9381 ve R2 test = 0.9404). Sonuçlar, geliştirilen modelimizin gıda endüstrisinde deneysel çalışmaların yerine hücre kültürü çalışmalarında başarılı bir şekilde kullanılabileceğini göstermektedir.
Investigation of Yeast Cells Life Cycle Parameters by Using Multi-Layer Perceptron Artificial Neural Network
Examining the growth parameters of yeast cells in the food industry causes to increase both time andlabor costs. Simulation models can be put forward to reduce these costs. In this study aimed that designa simulation model for growth cycle parameters of Saccharomyces cerevisiae by using the Multi-LayerPerceptron Neural Network (MLPNN). While cultivation time is defined as input parameter in thismodel, the cell count per hour and growth rate is determined as output parameters. In the designedmodel, two hidden layer back propagation neural networks are preferred. The first hidden layer uses10 nodes, while the second hidden layer uses 2 nodes. For the training of this model, 144 experimentaldata are used, whereas 72 of these experimental data were used for testing the trained model. Thedeveloped model showed a high correlation on the growth curve and growth rate in the process bothtraining and test (R2training=0.9993 and R2test=0.9993 for growth curve; R2training=0.9381 and R2test=0.9404for growth rate). The results show that developed model can be used successfully in cell culture studiesinstead of experimental studies in food industry.
___
- Dogan, A. and Gonullu, G., 2018. Current Debates in
Sustainable Architecture Urban Design &
Environmental Studies, (Chapter Editors: Kara, Ç.,
Özden, G., Karacasu, M., Sarıçiçek, İ.: Artificial
Neural Network Modelling for Mode Choice
Prediction in Transportation Planning), IJOPEC
Publication Limited, İstanbul, 102-110.
- Garcia-Camacho, F., Lopez-Rosales, L., Sanchez-Miron,
A., Belarbi, E.H., Chisti, Y. and Molina-Grima, E.,
2016. Artificial neural network modeling for
predicting the growth of the microalga Karlodinium
veneficum, Algal Research, 14, 58-64.
- Garcia-Gimeno, R.M., Hervas-Martinez, C., de Siloniz,
M. I., 2002. Improving artificial neural networks
with a pruning methodology and genetic algorithms
for their application in microbial growth prediction
in food. International Journal of Food Microbiology,
72, 9–30.
- Gulbandilar, E., 1996. Effects of pulsing
electromagnetic field on the growth of
saccharomyces cerevisiae. (Master Thesis, in
Turkish), Osmangazi University, Institute of Health
Sciences, Eskisehir, Turkey, 48.
- Gulbandilar, E., 2005. Effects of pulsing
electromagnetic field on the growth of
Saccharomyces cerevisiae. Journal of Science and
Technology of Dumlupinar University, 9, 55-64.
- Gulbandilar, E. and Gulbandilar, A., 2016. Modeling of
Growth Curve of Saccharomyces cerevisiae yeast
Cell with ANFIS, in 4th International Symposium on
Innovative Technologies in Engineering and Science
3-5 November 2016 (ISITES2016), 2016,
Alanya/Antalya-Turkey.
Gurgun, V. and Halkman, A.K., 1990. Counting Methods
in Microbiology, 2nd ed.. Gıda Teknolojisi Yayınları,
Ankara, (In Turkish).
- Hajmeer, M.N., Basheer, L.A., Marsden J.L. and Fung,
D.Y.C., 2000. New Approach for Modeling
Generalized Microbial Growth Curves Using
Artificial Neural Networks. Journal of Rapid
Methods and Automation in Microbiology, 8, 265-
283.
- Hajmeera, M. and Basheer, I., 2002. A probabilistic
neural network approach for modeling and
classification of bacterial growth/no-growth data.
Journal of Microbiological Methods, 51, 217-226.
- Jeyamkondan , S., Jayas, D.S. and Holley. R.A., 2001.
Microbial growth modelling with artificial neural
networks. International Journal of Food
Microbiology, 6, 343–354.
- Karakus, N., 2019. Modeling of Growth Curve of Yeast
Cells with Expert Systems. (Master Thesis, in
Turkish), Kutahya Dumlupinar University, Institute
of Science and Technology, Kutahya, Turkey, 75.
- Pappu, J.S.M., Vijayakumar, G.K. and Ramamurthy, V.,
2013. Artificial neural network model for predicting
production of Spirulina platensisin outdoor culture.
Bioresource Technology, 130, 224–230.
- Simon, L. and Kariml, M. N., 2001. Probabilistic neural
networks using Bayesian decision strategies and a
modified Gompertz model for growth phase
classification in the batch culture of Bacillus subtilis.
Biochemical Engineering Journal, 7, 41-48.
- Vasquez, N., Magan, C., Oblitas, J., Chuquizuta T.,
Avila-George, H. and Castro, W., 2018. Comparison
between artificial neural network and partial least
squares regression models for hardness modeling
during the ripening process of Swiss-type cheese
using spectral profiles. Journal of Food Engineering,
219, 8-15.
- Veysa O., Elias, S.O., Sampersa, I. and Tondo, E.C.,
2016. Modelling the growth of Salmonella spp. and
Escherichia coli O157 on lettuce. Procedia Food
Science, 7, 168-172.
- 1-Cerevisiae.S.(2017)
https://en.wikipedia.org/wiki/Saccharomyces_cere
visiae (04.2017)