ibm-research/otter_primekg

---
license: mit
---

# Otter PrimeKG Dataset Card
The Otter PrimeKG dataset contains 12,757,257 triples with Proteins, Drugs and Diseases. It contains protein sequences, SMILES and text

## Dataset details
#### PrimeKG
PrimeKG (the Precision Medicine Knowledge Graph) integrates 20 biomedical resources, it describes 17,080 diseases with 4 million relationships. PrimeKG includes nodes describing Gene/Proteins (29,786) and Drugs (7,957 nodes). The Multimodal Knowledge Graph (MKG) that we built from PrimeKG contains 13 modalities, 12,757,300 edges (154,130 data properties, and 12,603,170 object properties), including 642,150 edges describing interactions between proteins, 25,653 edges describing drug-protein interactions, and 2,672,628 describing interactions between drugs.

**Original dataset:**
- [GitHub Repo](https://zitniklab.hms.harvard.edu/projects/PrimeKG)
- Citation: Chandak, P., Huang, K. & Zitnik, M. Building a knowledge graph to enable precision medicine. Sci Data 10, 67 (2023). https://doi.org/10.1038/s41597-023-01960-3

**Paper or resources for more information:**
- [GitHub Repo](https://github.com/IBM/otter-knowledge)
- [Paper](https://arxiv.org/abs/2306.12802)

**License:**

MIT

**Where to send questions or comments about the dataset:**
- [GitHub Repo](https://github.com/IBM/otter-knowledge)

**Models trained on Otter PrimeKG**
- [ibm/otter_primekg_classifier](https://huggingface.co/ibm/otter_primekg_classifier)
- [ibm/otter_primekg_distmult](https://huggingface.co/ibm/otter_primekg_distmult)
- [ibm/otter_primekg_transe](https://huggingface.co/ibm/otter_primekg_transe)