GA4GH Phenopacket to LLM prompt
Prompts generated with this code base have been used for:
- A benchmark of some large language models against a differential diagnostics tool, Exomiser (code here). Our evidence shows Exomiser outperforms the LLMs as per our preprint, data at this zenodo.
- A comparison of gpt-4o's and Meditron3-70B ability to carry out differential diagnosis when prompted in 10 different languages, results published in eBioMedicine and at zenodo.
At this related link for you can find 5K+ patient descriptions in 10 languages created from these phenopackets, automatically updated when a new release of the phenopacket-store happens.