Project 4: Deep learning of molecular graphs

15. December 2021

The aim of this project is to develop more informative representations and machine learning models of atomically resolved chemical and biological systems. These representations and models will take into account local as well as non-local information about the structure of the system. These innovations will be used in a variety of other projects in the SIMPLAIX consortium.

The project is hosted by the labs of Fred Hamprecht and Pascal Friederich who have complementary experience in developing and applying machine learning models for applications in medicine, chemistry, and materials science.

Graph neural networks are used to predict molecular properties by extracting representations directly from molecular graphs.

Team

Prof. Dr. Fred Hamprecht

Principal Investigator (Heidelberg University)

Phone: +49 6221 5414803

More Information

T.T.-Prof. Dr. Pascal Friederich

Principal Investigator (Karlsruhe Institute of Technology)

Phone: +49 721 60844764

More Information

Switch to the German homepage or stay on this page