Use of a two-phase aqueous system with a deep eutectic solvent in recovery and purification: modeling and simulation with Lifted Relational Neural Networks
beta-galactosidase; deep eutectic solvents; neural networks; Lifted Relational Neural Networks
There are a growing number of successful applications being developed with Relational Deep Learning. Among them, the identification of side effects of polymedication, discovery of antibiotics, among others. This great success in obtaining practical results makes this technique attractive for the industry, which seeks to simulate the systems that are in its production chain aiming at optimization. One of the most important steps in the development of bioprocesses refers to the recovery and purification of biomolecules. In this context, the present study proposes to develop an effective methodology for building Relational Deep Learning models that can be used in the development of complex bioprocesses, at different scales, focusing on the recovery and purification steps of biomolecules.