Prediction of Rumen Fungal Xylanase Enzyme Activity with Python Based Fuzzy Logic Model: Analysis of Microbial Production Conditions

Öz

The xylanase enzyme, belonging to the rumen fungal enzyme family, plays a crucial role in the degradation of lignocellulosic structures. Moreover, this enzyme has been effectively utilized in various industrial applications after being isolated from a wide range of organisms. However, the sensitivity of enzyme activity to environmental conditions complicates the determination of optimal production and application parameters, thereby increasing the need for computational approaches. In this study, an artificial intelligence-assisted approach was adopted to predict and model the activity of xylanase. In this study, it was aimed to give the programming language in a differentiated way, in other words, to explain the tasks of the commands given in order to better understand the rumen microbial based enzyme analyses to be performed in the future. In this context, the Python programming language was selected due to its extensive capabilities in data analysis and algorithmic modeling. Through the use of Python’s powerful scientific libraries (such as NumPy, Pandas, Scikit-Fuzzy, and Matplotlib), a fuzzy logic method—well-suited to modeling the uncertain and nonlinear nature of biological systems—was implemented. During the modeling process, temperature, pH, and substrate concentration were selected as the key environmental parameters affecting xylanase activity. Instead of being defined with traditional sharp boundaries, these variables were classified into "low", "medium", and "high" categories using fuzzy sets. With the help of predefined membership functions, the system’s responses to these parameters were analyzed through fuzzy logic rules. In this way, a flexible model capable of evaluating not only the optimal conditions but also variations under boundary conditions was developed. As a result, the Python-based fuzzy logic model successfully predicted the activity levels of xylanase under different environmental conditions. It was determined that the enzyme exhibited optimal activity within the pH range of 5.5–6.5, the temperature range of 50–60 °C, and at a substrate concentration of approximately 5 U/mL. When compared with the literature, these findings suggest that fuzzy logic can serve as an alternative method in the digitalization and optimization of microbial enzyme production processes. In conclusion, the fuzzy logic approach supported by Python demonstrates significant potential in modeling biotechnological processes, particularly in interpreting the uncertainties inherent in biological data. This study thus provides a valuable reference for future computational modeling applications in enzyme technology and bioprocess engineering.

Anahtar Kelimeler

• Xylanase
• Python
• Fuzzy
• Fuzzy Logic
• Neocallimastix

Python Tabanlı Bulanık Mantık Modeli ile Rumen Fungal Ksilanaz Enzim Aktivitesinin Tahmini: Mikrobiyal Üretim Koşullarının Analizi

Öz

Rumen fungal enzim ailesine ait olan ksilanaz enzimi, lignoselülozik yapının parçalanmasında önemli bir rol oynamaktadır. Ayrıca, bu enzim günümüzde çeşitli canlı gruplarından izole edilerek endüstriyel alanda etkin bir şekilde kullanılmaktadır. Ancak enzim aktivitesinin çevresel koşullara duyarlılığı, optimum üretim ve uygulama koşullarının belirlenmesini zorlaştırmakta ve bu noktada hesaplamalı yaklaşımlara olan ihtiyaç artmaktadır. Bu çalışmada, ksilanaz enziminin aktivitesini tahmin etmek ve modellemek amacıyla yapay zekâ destekli bir yaklaşım benimsenmiş; bu bağlamda, veri analizi ve algoritmik modelleme açısından geniş olanaklar sunan Python programlama dili tercih edilmiştir. Çalışmada proglama dilinin ayırtılı bir şekilde verilmesi başka bir ifade ile verilen komutların görevlerinin açıklanması gelecekte yapilacak olan rumen mikrobiyal temmelli enzim analizlerinin daha iyi bir sekilde anlaşılması hedeflenmiştir. Python’un bilimsel hesaplamalara yönelik güçlü kütüphaneleri (örneğin NumPy, Pandas, Scikit-Fuzzy, Matplotlib) aracılığıyla, biyolojik sistemlerin belirsiz ve doğrusal olmayan doğasını modellemeye elverişli bir yöntem olan bulanık mantık yaklaşımı uygulanmıştır. Modelleme sürecinde, ksilanaz enziminin aktivitesini etkileyen temel çevresel parametreler olarak sıcaklık, pH ve substrat konsantrasyonu seçilmiştir. Bu değişkenler, klasik keskin sınırlarla tanımlanmak yerine, bulanık kümeler aracılığıyla "düşük", "orta" ve "yüksek" olarak sınıflandırılmıştır. Belirlenen üyelik fonksiyonları yardımıyla, sistemin bu parametrelere karşı verdiği tepkiler bulanık mantık kuralları doğrultusunda analiz edilmiştir. Böylece, yalnızca optimum koşulları değil, aynı zamanda sınır koşullarındaki varyasyonları da değerlendirebilen esnek bir model ortaya konmuştur. Sonuç olarak, Python destekli bulanık mantık modeli sayesinde, ksilanaz enziminin farklı çevresel koşullardaki aktivite düzeyleri başarıyla öngörülmüş; bu enzimin optimal etkinlik gösterdiği pH aralığının 5.5-6.5, sıcaklık aralığının 50-60 °C ve substrat konsantrasyonunun yaklaşık 5 U/mL aralığında olduğu belirlenmiştir. Bu sonuçlar literatür ile karşılaştırıldığında bulanık mantığın, mikrobiyal enzim üretim süreçlerinin dijitalleştirilmesi ve optimizasyonunda alternatif bir yol olarak kullanılabileceği önerilmiştir. Elde edilen bulgular, biyoteknolojik süreçlerin modellenmesinde bulanık mantığın, özellikle biyolojik verilerin doğasında bulunan belirsizlikleri anlamlandırma noktasında önemli bir potansiyele sahip olduğunu ortaya koymaktadır.

Anahtar Kelimeler

• Ksilanaz
• Python
• Fuzzy
• Bulanık Mantık
• Neocallimastix

Kaynakça

1. Badar, M., Ahmad, I., Mir, A. A., Ahmed, S., and Waqas, A. (2022). An autonomous hybrid DC microgrid with ANN-fuzzy and adaptive terminal sliding mode multi-level control structure. Control Engineering Practice, 121, 105036.
2. Cai, S., Li, J., Hu, F. Z., Zhang, K., Luo, Y., Janto, B., ... & Dong, X. (2010). Cellulosilyticum ruminicola, a newly described rumen bacterium that possesses redundant fibrolytic-protein-encoding genes and degrades lignocellulose with multiple carbohydrate-borne fibrolytic enzymes. Applied and environmental microbiology, 76(12), 3818-3824.
3. Çalışkan, R., Yağcı, S., & Kocabaş, D. S. (2014). Lignoselülozik biyokütlenin ekstrüzyon ile ön işlemi ve enzimatik yolla ksilooligosakkarit üretimi. Karamanoğlu Mehmetbey Üniversitesi Mühendislik ve Doğa Bilimleri Dergisi, 1(1), 156-171.
4. Cao, S., Zeng, Y., Yang, S., and Cao, S. (2021). Research on Python data visualization technology. In Journal of physics: Conference series (Vol. 1757, No. 1, p. 012122). IOP Publishing.
5. Chuensiri, S., Katchasuwanmanee, K., Wisessint, A., Jotisankasa, A., Soralump, C., Siriyakorn, V., ... and Sanposh, P. (2024). Implementation of Adaptive Network-Based Fuzzy Inference for Hybrid Ground Source Heat Pump. IEEE Access.
6. Comlekcıoglu, U., FC, Y., Keser, S., BM, K., Battaloglu, G., & Ozkose, E. (2012). Neocallimastix sp. ve Orpinomyces sp.’nin Enzim Üretimleri Üzerine Karbon Kaynaklarının Etkisi. Kafkas Üniversitesi Veteriner Fakültesi Dergisi, 18(5).
7. Daole, M., Schiavo, A., Bárcena, J. L. C., Ducange, P., Marcelloni, F., and Renda, A. (2023). OpenFL-XAI: Federated learning of explainable artificial intelligence models in Python. SoftwareX, 23, 101505.
8. Daway, H. G., Daway, E. G., and Kareem, H. H. (2020). Colour image enhancement by fuzzy logic based on sigmoid membership function. International Journal of Intelligent Engineering and Systems, 13(5), 238-246.

9. Deng, J., and Deng, Y. (2021). Information volume of fuzzy membership function. International Journal of Computers Communications & Control, 16(1).
10. Deng, Z., Wang, Y., Mao, J., and Ye, M. (2017). Investigating the Relationship between the Substrates’ Consumption and Their Abundances in a Complex Enzymatic System. Analytical chemistry, 89(20), 10644-10648.
11. Dong, Y., Li, G., Jiang, X., and Jin, Z. (2023). Antecedent Predictions Are More Important Than You Think: An Effective Method for Tree-Based Code Generation. In ECAI 2023 (pp. 565-574). IOS Press.
12. Dos Santos, J. P., da Rosa Zavareze, E., Dias, A. R. G., & Vanier, N. L. (2018). Immobilization of xylanase and xylanase–β-cyclodextrin complex in polyvinyl alcohol via electrospinning improves enzyme activity at a wide pH and temperature range. International Journal of Biological Macromolecules, 118, 1676-1684.
13. Gajare, A., Tembe, Y., Sawant, S., and Mangala, R. (2021, October). CircuitScribe: Block Diagram based Circuit Simulation Application. In 2021 IEEE Mysore Sub Section International Conference (MysuruCon) (pp. 492-497). IEEE.
14. Gilmiyarov, R. V., Galimzyanova, L. I., Porshnev, S., and Chernova, O. (2023, May). Comparative Quality Analysis of Random Number Generators in MATLAB and NumPy packages. In 2023 IEEE Ural-Siberian Conference on Biomedical Engineering, Radioelectronics and Information Technology (USBEREIT) (pp. 293-297). IEEE.
15. Grum, M., Kotarski, D., Ambros, M., Biru, T., Krallmann, H., and Gronau, N. (2021). Managing Knowledge of Intelligent Systems: The Design of a Chatbot Using Domain-Specific Knowledge. In Business Modeling and Software Design: 11th International Symposium, BMSD 2021, Sofia, Bulgaria, July 5–7, 2021, Proceedings 11 (pp. 78-96). Springer International Publishing.
16. Harris, C. R., Millman, K. J., Van Der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., ... and Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825), 357-362.
17. Htun, S. N. N., Egami, S., and Fukuda, K. (2024). Activity scenarios simulation by discovering knowledge through activities of daily living datasets. SICE Journal of Control, Measurement, and System Integration, 17(1), 87-105.
18. Hu, J., Arantes, V., & Saddler, J. N. (2011). The enhancement of enzymatic hydrolysis of lignocellulosic substrates by the addition of accessory enzymes such as xylanase: is it an additive or synergistic effect?. Biotechnology for biofuels, 4, 1-14.
19. Jianling, W., Xian, Y., Nan, L., and Detong, C. (2023, July). Fuzzy Control Strategy for Energy Allocation Optimization in Extended Range Mode. In 2023 8th International Conference on Image, Vision and Computing (ICIVC) (pp. 847-852). IEEE.
20. Jung, H. G., & Casler, M. D. (2006). Maize stem tissues: impact of development on cell wall degradability. Crop science, 46(4), 1801-1809.
21. Kar, B., & Torcan, B. (2023). Isolation, morphological identification, and xylanase characteristics of anaerobic gut fungi Neocallimastix from Anatolian wild goat. Journal of Basic Microbiology, 63(3-4), 377-388.
22. Khandeparker, R., & Numan, M. T. (2008). Bifunctional xylanases and their potential use in biotechnology. Journal of Industrial Microbiology and Biotechnology, 35(7), 635-644.
23. Lin, Y. H., Yu, C. M., and Wu, C. Y. (2021). Towards the design and implementation of an image-based navigation system of an autonomous underwater vehicle combining a color recognition technique and a fuzzy logic controller. Sensors, 21(12), 4053.
24. Liu, J. C., Cheng, H. L., Lai, Y. H., Hu, C. Y., and Chen, Y. C. (2024). A fragment of the β-glucosidase gene from the rumen fungus Neocallimastix patriciarum J11 encodes a recombinant protein that exhibits activities in β-glucosidase and β-glucanase. Biochemical and Biophysical Research Communications, 732, 150406.
25. Ma, H., Zheng, W., Wang, T., and Zhang, P. (2024, May). Performance improvement for Python-based Dynamic System Simulation. In 2024 3rd International Conference on Energy, Power and Electrical Technology (ICEPET) (pp. 1432-1435). IEEE.
26. May, R. M., Goebbert, K. H., Thielen, J. E., Leeman, J. R., Camron, M. D., Bruick, Z., ... and Marsh, P. T. (2022). MetPy: A meteorological Python library for data analysis and visualization. Bulletin of the American Meteorological Society, 103(10), E2273-E2284.
27. Mountfort, D. O., & Asher, R. A. (1989). Production of xylanase by the ruminal anaerobic fungus Neocallimastix frontalis. Applied and Environmental Microbiology, 55(4), 1016-1022.
28. Raja, K. (2023). Python-based fuzzy logic in automatic washer control system. Soft Computing, 27(10), 6159-6185.
29. Pant, S., Prakash, A., Vundavilli, P. R., Khadanga, K. C., Kuila, A., Aminabhavi, T. M., and Garlapati, V. K. (2024). Concomitant inhibitor-tolerant cellulase and xylanase production towards sustainable bioethanol production by Zasmidiumcellare CBS 146.36. Fuel, 375, 132593.
30. Papa, G., Varanasi, P., Sun, L., Cheng, G., Stavila, V., Holmes, B., ... & Singh, S. (2012). Exploring the effect of different plant lignin content and composition on ionic liquid pretreatment efficiency and enzymatic saccharification of Eucalyptus globulus L. mutants. Bioresource Technology, 117, 352-359.
31. Samsingh, V., Ramachandran, A., Selvam, A., and Subramanian, K. (2021). Python implementation of fuzzy logic for artificial intelligence modelling and analysis of important parameters in drilling of hybrid fiber composite (HFC). In IOP Conference Series: Materials Science and Engineering (Vol. 1012, No. 1, p. 012037). IOP Publishing.
32. Singhala, P., Shah, D., and Patel, B. (2014). Temperature control using fuzzy logic. arXiv preprint arXiv:1402.3654. Spolaor, S., Fuchs, C., Cazzaniga, P., Kaymak, U., Besozzi, D., and Nobile, M. S. (2020). Simpful: a user-friendly Python library for fuzzy logic. International Journal of Computational Intelligence Systems, 13(1), 1687-1698.
33. Stabel, M., Hagemeister, J., Heck, Z., Aliyu, H., and Ochsenreither, K. (2021). Characterization and Phylogenetic Analysis of a Novel GH43 β-Xylosidase From Neocallimastix californiae. Frontiers in fungal biology, 2, 692804.
34. Steubing, B., de Koning, D., Haas, A., and Mutel, C. L. (2020). The Activity Browser—An open source LCA software building on top of the brightway framework. Software Impacts, 3, 100012.
35. Trevizano, L. M., Ventorim, R. Z., de Rezende, S. T., Junior, F. P. S., & Guimarães, V. M. (2012). Thermostability improvement of Orpinomyces sp. xylanase by directed evolution. Journal of Molecular Catalysis B: Enzymatic, 81, 12-18.
36. Tura, A. D., Lemu, H. G., Mamo, H. B., and Santhosh, A. J. (2023). Prediction of tensile strength in fused deposition modeling process using artificial neural network and fuzzy logic. Progress in Additive Manufacturing, 8(3), 529-539.
37. Wang, Y., & McAllister, T. A. (2002). Rumen microbes, enzymes and feed digestion-a review. Asian-Australasian Journal of Animal Sciences, 15(11), 1659-1676.
38. Zhang, M., Qiu, Q., Zhao, X., Ouyang, K., & Liu, C. (2024). Characterization of Novel Multifunctional Xylanase from Rumen Metagenome and Its Effects on In Vitro Microbial Fermentation of Wheat Straw. Fermentation, 10(11), 574.
39. Zhang, Y., Yang, H., Yu, X., Kong, H., Chen, J., Luo, H., ... and Yao, B. (2019). Synergistic effect of acetyl xylan esterase from Talaromyces leycettanus JCM12802 and xylanase from Neocallimastix patriciarum achieved by introducing carbohydrate-binding module-1. AMB Express, 9, 1-12.
40. Zhu, D., Liu, X., Xie, X., Yang, S., Lin, H., and Chen, H. (2020). Characteristics of a XIP‐resistant xylanase from Neocallimastix sp. GMLF 1 and its advantage in barley malt saccharification. International Journal of Food Science & Technology, 55(5), 2152-2160.