Establishing a knowledge graph community in biomedical science

• Applicants: Dr. Sebastian Lobentanzer; Dr. Inga Ulusoy
• Subject Area: Medical Informatics and Medical Bioinformatics
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 528753569

Many modern methods in the biomedical sciences require - or at least benefit - from using prior knowledge, for instance about genes, proteins, and diseases. Knowledge graphs (KGs; i.e., knowledge representations in the form of a machine-readable graph structure) are rapidly becoming the dominant solution for storing, manipulating, and sharing this prior knowledge. With OmniPath and its infrastructure software Pypath, we provide a well-used resource of knowledge around molecular signalling to the community. In addition, we have recently started to generalise and standardise the process of biomedical knowledge graph creation with our BioCypher software. With the present proposal, we aim to increase the reach and usefulness of our methods, and actively build a community of users and developers in Germany and abroad. The proposed work will increase the usability and longevity of our knowledge-providing software solutions. Success of research software in a rapidly evolving environment depends on continuous adaptation of its design. Simultaneously, the longevity of research software in the face of short-term contracts for most of the active developers depends on the cultivation of an active developer community. Our proposal is a refactoring plan that shall secure our software’s ability to support the research community’s modelling and data analysis activities despite their increasing demands thanks to evolving omics technologies. Through increased usability, we want to increase the user-base, which should ultimately enable the recruitment of community developers to achieve a self-sustaining community. The combination of a well-used, established resource (OmniPath) and a novel FAIR and TRUSTworthy framework (BioCypher) puts us in a unique position of being able to increase the reusability of knowledge provision software by a large margin. Compared to the current landscape of siloed low-reusability solutions, the present project can yield large inclusivity and productivity benefits to the wider field of biomedical research.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr.-Ing. Julio Saez-Rodriguez