Investigate indication linking #13

Closed
opened 4 years ago by youainti · 5 comments
youainti commented 4 years ago (Migrated from gitea.kgjk.icu)

I know of the following resources:

  • MeSH
  • Snowmed
  • RXnorm

I need to figure out how to

  • Link Trials to indications
  • Link Drugs to indications
  • Group indications

This is going to take a bit.

This issue is to handle a discussion with myself on the topic about what needs to happen.
The actual things I need to do will be handled in other issues.

I know of the following resources: - MeSH - Snowmed - RXnorm I need to figure out how to - Link Trials to indications - Link Drugs to indications - Group indications This is going to take a bit. This issue is to handle a discussion with myself on the topic about what needs to happen. The actual things I need to do will be handled in other issues.
youainti commented 3 years ago (Migrated from gitea.kgjk.icu)

I have a bit more written down in logseq. there is a WHO taxonomy used by other papers.

I have a bit more written down in logseq. there is a WHO taxonomy used by other papers.
youainti commented 3 years ago (Migrated from gitea.kgjk.icu)

so, the clinicla trials indications are indexed in mesh terms by the table ctgov.browse_conditions.

so, the clinicla trials indications are indexed in mesh terms by the table `ctgov.browse_conditions`.
youainti commented 3 years ago (Migrated from gitea.kgjk.icu)

I need a way to link spls indications.

I need a way to link spls indications.
youainti commented 3 years ago (Migrated from gitea.kgjk.icu)

I currently have the following data links outlined

  flowchart LR
  
  
  %% Nodes
  spl[(SPLs describe drugs on market)]
  rxnorm[[RxNorm]]
  dom([drugs on market]):::varsOfInterest
  ct[(ClinicalTrials History)]  
  subgraph mesh[Mesh Taxonomy]
    meshd[[MeSH drugs]]
    meshi[[MeSH indications]]
  end
  who[[WHO ATC]]
  
  %% Class definitions
  classDef varsOfInterest fill:#959
  
  meshd <==RxNorm provides an interface to MeSH ==> rxnorm
  spl -.  rxnorm_mapping file .-> rxnorm
  spl -- spl all drug labels --> dom
  ct -. ctgov.browse_interventions .-> meshd
  ct -. ctgov.browse_conditions .-> meshi
  who <==I believe there is an interface==> mesh
  • I am missing how to link indications to drugs
    • there may be the option to do through the WHO ATC. As this provides automatic groups I'd probably go with this.
    • Or I can try to group the trials through ctgov.browse_conditions table and group them. I doubt I have the ability to do this well.
    • There might be a hybrid approach where you match MeSH terms from trials to WHO ATC and check for poor matches.
I currently have the following data links outlined ``` mermaid flowchart LR %% Nodes spl[(SPLs describe drugs on market)] rxnorm[[RxNorm]] dom([drugs on market]):::varsOfInterest ct[(ClinicalTrials History)] subgraph mesh[Mesh Taxonomy] meshd[[MeSH drugs]] meshi[[MeSH indications]] end who[[WHO ATC]] %% Class definitions classDef varsOfInterest fill:#959 meshd <==RxNorm provides an interface to MeSH ==> rxnorm spl -. rxnorm_mapping file .-> rxnorm spl -- spl all drug labels --> dom ct -. ctgov.browse_interventions .-> meshd ct -. ctgov.browse_conditions .-> meshi who <==I believe there is an interface==> mesh ``` - I am missing how to link indications to drugs - there may be the option to do through the WHO ATC. As this provides automatic groups I'd probably go with this. - Or I can try to group the trials through `ctgov.browse_conditions` table and group them. I doubt I have the ability to do this well. - There might be a hybrid approach where you match MeSH terms from trials to WHO ATC and check for poor matches.
youainti commented 3 years ago (Migrated from gitea.kgjk.icu)

ok, investigated.

ok, investigated.
Sign in to join this conversation.
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: youainti/ClinicalTrialsDataProcessing#13
Loading…
There is no content yet.