Save the date!

2nd July 2020, 9:00 – 17:00, Schloss Birlinghoven

Fraunhofer Symposium

Knowledge graphs are way to represent complex biomedical knowledge in a structured, searchable and interpretable form. In addition, there is a high interest in AI based approaches across the entire biopharmaceutical value chain. More recently, AI methods integrating knowledge graphs and experimental lab data have been suggested.

The purpose of the symposium is to get an overview on the latest trends on

• Using knowledge graphs & network algorithms to build disease specific models in the context of neurodegenerative diseases with focus on Alzheimer’s disease
• Development of novel strategies for high throughput and high content screening & experimental validation of in silico hypotheses with the help of knowledge graphs
• state-of-the-art algorithms and text mining software for large scale document processing as a basis for building knowledge graph
• AI based approaches for integrating data and knowledge graphs for Drug Discovery.

Preliminary Programme

• Prof. Dr. Martin Hofmann-Apitius: setting the scene
   

Keynote:

• To be confirmed
   

Computational Modeling Strategies to Aid Hypothesis Generation

• Prof. Dr. Holger Fröhlich: "Using Knowledge Graphs and AI for target prioritization and prediction of adverse events"
• Dr. Marc Jacobs: "The Human Brain Pharmacome – a quantitative mechanistic knowledge graph"
• Dr. Joachim Köhler: The Human Brain Pharmacome – large scale state of the art table extraction from PDF
   

In vivo detection of modulation of mechanisms - finding the right endpoints

• Dr. Ole Pless: "Biological space: Validation of in silico predicted modulators of Alzheimer´s disease in physiologically relevant cell models based on pluripotent stem cells"
  
• Dr. Phil Gribbon: "Chemical space: New strategies for high throughput screening and cooperation models with the pharmaceutical industry"
  
• Dr. Stephan Brock: "Drug repurposing – How we did it for Gates Foundation"
   

Summary

• Dr. Carsten Claussen: "lessons learned"
   

Hands-on sessions

• Large scale table extraction and retrieval of quantitative data
  
• Deep learning and interpretable models for compound and target selection
  
• Querying mechanistic knowledge graphs