CD Laboratory for Molecular Informatics in the Biosciences

Artificial intelligence and simulation methods are playing an increasingly important role in the biosciences. Today, they are used in practically all phases of drug discovery and optimisation. They are used, for example, to predict the bioactivity spectra of drug candidates, which can provide information about pharmacological efficacy and toxicity, or the distribution of agrochemicals in crops, which allows conclusions to be drawn about the efficacy of the substances. Large, high-quality data sets are required for the further development of computer-based approaches. These are difficult to access for academic research.

With this CD Laboratory, several expert groups and institutions are joining forces to push the boundaries of machine learning and molecular dynamics simulations for the prediction of the chemical and biological properties of small organic molecules and proteins. Specifically, innovative methods for the prediction of bioactivity, metabolism and toxicity will be developed. The large amounts of data available as a result of the collaboration play a central role here. Great importance is attached to the predictive power, applicability and interpretability of the approaches. The methods and computer models developed by this CD Laboratory are integrated into a consistent, maintainable and scalable software platform. Essential modules of this platform will be made available for scientific research as open source applications.