Add pipeline for ingesting from Molecular Case Studies (MCS) papers

monarch-initiative / phenopacket-store

Collection of phenopackets

https://monarch-initiative.github.io/phenopacket-store/

BSD 3-Clause "New" or "Revised" License

12 stars 4 forks source link

Add pipeline for ingesting from Molecular Case Studies (MCS) papers #85

Open cmungall opened 3 months ago

cmungall commented 3 months ago

E.g

https://molecularcasestudies.cshlp.org/content/9/4/a006294.full

All MCS case reports have HPO IDs pre-annotated.

Of course manual curation is required to capture the individuals but it could still be useful to have a placeholder phenopacket as the g2p relationship is potentially useful

pnrobinson commented 3 months ago

It would also be great to see if we can develop an LLM powered tool to decipher tables such as this: https://molecularcasestudies.cshlp.org/content/9/4/a006294/T1.expansion.html We also need to extract the variants. I think that Z. Liu at NLM had a good tool for that. Love this idea, maybe we can write a grant proposal for it.

cmungall commented 2 months ago

https://chat.openai.com/share/69dd5d8d-dc24-49bf-8ad3-68484504839d

(may not be visible if you're not a subscriber)

It should be possible to recapitulate this in openinterpreter

pnrobinson commented 2 months ago