Update/Create SQL rules for integrated diagnosis, broad, and short histologies

[jharenza]

What data file(s) does this issue pertain to?

pbta-histologies.tsv

What release are you using?

will use V17 from #732

Put a link to the relevant section of the OpenPBTA-manuscript here.

https://alexslemonade.github.io/OpenPBTA-manuscript/#clinical-data-harmonization

Put your question or report your issue here.

@baileyckelly or @aadamk, correct me if I am wrong:

This is an overview of the new workflow for creating the pbta-histologies.tsv file:

integrated_diagnosis:

Using updated pathology_diagnosis values:

1. Generate updated integrated_diagnosis using molecular subtyping modules in openPBTA_analysis repo, later to be implemented in d3b analysis toolkit (@kgaonkar6)
2. Then run OpenPBTA pathology module to update integrated_diagnosis using new pathological diagnoses, also part of the openPBTA_analysis repo as in (1) and to be implemented in the d3b analysis toolkit (@kgaonkar6)
3. Update CBTTC 'other' to new integrated_diagnosis if there are continued updates in the data warehouse (@baileyckelly)
4. For all remaining, leverage original pathology_diagnosis and populate as integrated_diagnosis (@baileyckelly)

broad_histology:

• Dependent on integrated_diagnosis
• Broad Histology: Broad 2016 WHO classification (https://doi.org/f8mspx, https://alexslemonade.github.io/OpenPBTA-manuscript/, https://link.springer.com/article/10.1007/s00401-016-1545-1/tables/1)
  • Process
    • Construct sql logic that maps integrated_diagnosis to WHO 2016 fields (@baileyckelly )

short_histology:

• Dependent on integrated_diagnosis
• Abbreviated version of integrated_diagnosis
  • Process
    • Construct sql logic that maps integrated_diagnosis to short_histology (@bailey Farrow)

Considerations

• From V16 prior, the workflow for broad_histology and short_histology depends on integrated_diagnosis, which depends on pathology_diagnosis and molecular_subtype.
• However, molecular_subtype modules currently depend on broad_histology, short_histology and/or integrated_diagnosis. So, we will have to rework these.

[jharenza]

I am going to close this, since we are utilizing an input file via molecular-subtyping-integrate instead.