At the intersection of open science and artificial intelligence (AI), we work to make biomedical research more accessible and AI methodologies more transparent and effective.
We enhance access to scientific knowledge and computational tools by developing open-source, user-friendly software for researchers across disciplines.
We critically examine artificial intelligence techniques—from basic linear models to advanced deep learning—to understand their strengths and limitations, guiding their effective application and improvement.
Important
We believe in slo science (pun intended), where we approach problems with a focus on understanding and sustainability. In today's research landscape, this can be counterintuitive; however, we believe that the individual's flexibility to explore without pressure is what makes science great.
Our core principles:
- 🔄 Reproducibility: If it's not reproducible, it's not science
- 🤖 Embracing AI: Understanding AI as a powerful tool while remaining critical
- 🏗️ Infrastructure: Automating the mundane to focus on innovation
- 📚 Accessibility: Making science available to everyone
We investigate what makes ML/DL models truly useful in biomedical contexts:
- 🏥 Patient classification
- 🧪 Perturbation effect prediction
- 🧬 Gene expression alignment
- ⛓️ Causality research
- 🔗 BioCypher - Modern knowledge graph framework
- 💬 BioChatter - LLM framework for biomedicine
A three-year DFG-funded initiative establishing a biomedical knowledge graph community, evolving from the OmniPath resource.
An Open Targets-funded project modernizing knowledge infrastructure:
- 📑 Text information extraction
- 🕸️ Knowledge graph representation
- 🤖 ML framework integration
- 🗣️ LLM accessibility enhancement
We believe in the power of open collaboration! Check out our open science guide to learn how we navigate the intersection of academic research and open-source development.
We actively contribute to these open-source projects:
| Project | Focus | Our Involvement |
|---|---|---|
| 🔗 BioCypher | Scientific Knowledge Management | Infrastructure, Tool Development & Documentation |
| 🎯 Open Targets | Drug Target Discovery | Platform Development & Knowledge Integration |
| 🐍 PyOpenSci | Scientific Python Ecosystem | Infrastructure, Package Reviews & Development |
| 📊 scverse | Single-Cell Analysis | Tool Development & Documentation |
| 🧬 Bioconductor | R Bioinformatics Framework | Accessibility & Cloud Infrastructure |
We are grateful for the support of:
- Saez-Rodriguez group
- Open Targets
- Heidelberg University Scientific Software Center
- ChEMBL
- Europe PMC
- EMBO
- The Francis Crick Institute
Making biomedical research more accessible, one commit at a time. ❤️