I saw a really interesting talk a few years ago by a clinician working at the Moore's Cancer Center that piqued my interest about it.

cells. There have been attempts with NTRK inhibitors to treat patients with an NTRK fusion. But there haven't been a lot of attempts to treat the WT form.

The process itself has been described in a few papers as an alt. metastatic route. So from that perspective alone, it deserves more study.

Yes. My mom described it as being stabbed by knitting needle. Cancer cells invade the nearby celiac nerve plexus causing the pain. There are 2 ways of treating it -- both of which are not durable. A celiac nerve block, where you attempt to kill nerve cells; and a palliative chemo to kill the cancer

As a comorbidity, it's always struck me as the red-headed stepchild of pancreatic cancer research. But it's the one thing that most patients will tell you affects their quality of life more than anything else. Alex Trebek said it was what he feared the most about the disease.

I wonder if perineural invasion was on the agenda?

The dataset metadata might point to the protocol repo, so that you could see how the data were generated, and what materials were used.

Thoughts? @atanas-g-atanasov.bsky.social @carolegoble.bsky.social @bffo.bsky.social

#FAIR #AI-ready

so that a column whose name is "Protein ID" actually references the "UniProt Accession" URI in the EDAM ontology, thus disambiguating it. Similar to RightField rightfield.org.uk

5) How would collaboration be supported within an institution? Across institutions? Between authors and reviewers?

6) How would annotations & provenance be supported in images, data, specialist file types, etc? You might want to tag fields with ontology refs...

4) Separating data from representation is critical. You want to be able to re-assemble the data in different ways. Paper => poster => presentation for example.

3) Working locally, you'll want everything in a bundle (like a zip file), when you publish it, then it can get taken apart and deposited where it needs to go. If you have to publish to multiple repos, things get complicated, and it's likely that something will go amiss.

Because of this, during the early stages of the process, data would be local and then reviewed at an institution before actual publication.

2) Either in academia or industry, scooping/IP protection is a concern. People will likely be working "offline" until they're ready to publish. In a way, this is a bit like they way git works, when you're working off your local repo prior to pushing it to the remote git server.

A few things to consider:
1) The design shows the destination - where data + papers + authors + images get deposited. How will it support the research, writing, and review steps that proceed publication?