Biotechnological processes play an important role in the course of raw material change in the chemical industry as well as in the pharmaceutical industry. Due to the complexity of the processes and the molecules involved, e.g. proteins, process design and optimization is often carried out empirically on the basis of high-throughput experiments, which is time-consuming and expensive. In preliminary work at LTD, the adsorption of proteins, which is the basis of a chromatographic separation - the most important purification process of proteins, was systematically investigated in experiments, and a model approach for the description of adsorption was developed. In this research project, a program for the simulation and prediction of elution profiles for the chromatographic separation of proteins is to be developed based on this model approach; a first prototype of the program is already available from preliminary work. The program is to be validated and tested on the basis of laboratory experiments and then used for model-based optimization of protein separations. Furthermore, a comprehensive analysis of protein adsorption data is to be carried out in cooperation with the Chair of "Visual Information Analysis" at the Department of Computer Science at TU Kaiserslautern. For this purpose, a first database from preliminary work can be used, which, however, will be extended within the scope of the research project. For this purpose, adsorption measurements will be carried out with a laboratory robot. The aim is to generate a comprehensive understanding of the factors influencing protein adsorption and to use this to develop novel data-driven models.

[back]