AICES - Data driven modelling in computational engineering sciences
Univ.-Prof. Dr.rer.nat. Andreas Schuppert
Field of Study: Modelling, Systems Biology, Bioinformatics
Main focus: Modelling of complex biologial systems, Reengineering of networks
Staff: 1 professor, 0 chief engineer, 4 scientific staff, 0 non-scientific staff, 0 stud. staff
Research Focus
Structured hybrid modeling:
Data driven modeling of complex systems with functional networks
Learning the behavior of highly complex systems from data plays a crucial role in a wide field of applications with high technological and economic potential:
Control and optimization of complex chemical plants with mixed reactive, separation and phase-transition Rational design of materials with new properties satisfying conflicting application requirements, based on known properties of compoundsIdentification of critical states in complex technical processesand many more. All these applications suffer from often inefficient learning technologies. Therefore our focus is the structural integration of knowledge with data driven modeling approaches: structured hybrid modelling
Reengineering of functional networks from data.
Structured hybrid modeling can be applied in inverse reengineering of functional networks from data. Solving the respective inverse problem allows insight into the underlying process on a coarse-grained level.
Multi-scale modeling of biological systems:
Modeling of regulation networks of biological systems
Identification and modeling of biological regulation networks is a key challenge for further progress in a wide range of biotechnological and medical application areas. Because of the tremendous complexity on the fully detailed level as well as the high redundancy of the data, structured hybrid modeling provides a promising approach for modeling biotechnological and biomedical processes.
Data driven modeling of biomedical systems from –omics data Prediction of biological phenotypes from –omics (genomics, proteomics, etc) data is challenging because of the n<<p problem: there are mostly much more parameters than data. Here we aim to reduce the complexity using hybrid models which are adapted to biological structures.
Contact
AICES - Data driven modelling in computational engineering science
Schinkelstr. 2
52062 Aachen
http://www.aices.rwth-aachen.de/
schuppert@aices.rwth-aachen.de
Phone: +49 (0)241 80 99144