OpenBioLink has concluded after an exciting and productive time! Here we provide entry points for discovering the project results. Of course, we will continue to maintain and improve OpenBioLink after the netidee project has ended.

Our project helps biomedical researchers to use AI and web technology to find connections in the network of biomedical research results more quickly, and thus to formulate and test promising hypotheses (e.g. 'drug X helps with disease Y').  On the other hand, our project helps AI developers to test and improve their models.

In the project, we also developed algorithms and tools for Explainable AI and Link Prediction in Knowledge Graphs that are generally applicable across domains.  🔗🔬🚀 

Entry points for discovering the project results

OpenBioLink (dataset) Github: https://github.com/OpenBioLink/OpenBioLink    

Link Explorer (User interface + framework) Github: https://github.com/OpenBioLink/Explorer  

SAFRAN (Link prediction algorithm that works together with Link Explorer) Github: https://github.com/OpenBioLink/SAFRAN  

Last but not least, our work and its result are document in two scientific papers:

Simon Ott, Adriano Barbosa-Silva, Matthias Samwald. “LinkExplorer: Predicting, Explaining and Exploring Links in Large Biomedical Knowledge Graphs”. Bioinformatics, February 2022, btac068. https://www.biorxiv.org/content/10.1101/2022.01.09.475537v2   

Simon Ott, Christian Meilicke and Matthias Samwald. “SAFRAN: An Interpretable, Rule-Based Link Prediction Method Outperforming Embedding Models”, 3rd Conference on Automated Knowledge Base Construction (AKBC 2021), 2021. https://openreview.net/forum?id=jCt9S_3w_S9 

We thank netidee for making these exciting developments possible! 🤓🤗