I currently develop computational methods for precision medicine in cancer research.
We have recently proposed PaccMann, an explainable deep learning model for in-silico screenings of anticancer drugs. PaccMann predicts IC50 drug sensitivity from a combination of a drug, gene expression data of the tumour and PPI. PaccMann t was demonstrated to surpass existing approaches for drug sensitivity prediction while leveraging model interpretability.
At the moment, we are creating a framework for the in-silico design of individualized anticancer drugs. Through deep generative model we want to tailor the de novo design of anticancer candidate drugs specifically to the biomolecular profile of a single patient, or a patient subgroup.
The pharma giant Roche has recently awarded the 1st price of the FXH Scientific Excellence Award 2019 to the project.
We are now seeking for collaborators for the experimental validation of the framework.
In 2017, Jannis obtained a BSc in Cognitive Science from University of Osnabrück (Germany), wherein he focused on computational neuroscience and conducted his thesis at the Oxford Centre for Theoretical Neuroscience and Artificial Intelligence (University of Oxford). Thereafter, he pursued a MSc in Neural Systems and Computation at ETH and UZH in Zurich, with a focus on computational psychiatry. After graduating with distinction in 2019, he continued his MSc thesis at the Computational Systems Biology group at IBM into his PhD.