A Soft Sensing Approach for Efficient Monitoring of Nanobody-Based Scorpion Antivenom Production
Document Type
Conference Paper
Year
2025
Authors
Juan C. Acosta-Pavas, David Camilo Corrales, Irene Martínez-Menéndez, Susana María Alonso Villela, Balkiss Bouhaouala-Zahar, Georgios K. Georgakilas, Konstantinos Mexis, Stefanos Xenios, Theodore Dalamagas, Antonis Kokosis, Michael O’donohue, Luc Fillaudeau & César A. Aceves-Lara
Source
Lecture Notes in Networks and Systems , 2025, 1259, pp. 271–281
Abstract
The growing demand for recombinant-based products has increased rapidly in recent years. One of the current market scopes is the production of recombinant scorpion antivenoms in E. coli. However, there are limitations in real-time quantification of Critical Quality Attributes (CQAs), such as recombinant proteins or biomass concentrations. Soft sensors merge as an alternative to solve this problem. In this work, six data-driven soft sensors were developed for online monitoring of the recombinant protein concentration. First, feature selection methods were tested to determine the relevant online variables to train the soft sensors. Then, the protein experimental dataset was augmented using a hybrid model. The K-Nearest Neighbors ), and the Radial Basis Function (RBF) Support Vector Machine (SVM) obtained the best results with values higher than 0.996, and values less than 0.058 and 0.051. However, the KNN soft sensor was computed in less time than the RBF SVM soft sensor. These results show the feasibility of KNN and RBF soft sensors for real-time monitoring of recombinant scorpion antivenoms.
