You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jun 21, 2023. It is now read-only.
The short_histology label as it currently is used contains a category called "Other", which includes tumor histologies of vastly different types, which can lead to improper grouping if it is used as the main label for a figure or analysis. For example, it includes "Benign tumors" (as labeled in broad_histology) such as Fibromas and Dermoid Cysts, but also "Metastatic secondary tumors", which I would imagine we would almost never want to group together.
We may want multiple "Other" categories, such as "Other-benign"?
I do not know if this may be resolved as part of #637, but any resolution will almost certainly result in breaking changes, not least for the color palettes in #622.
cc-ing @baileyckelly and @allisonheath to weigh in on this! As I am looking over the figures, I am thinking perhaps instead of Other, we may want to add short_histology for these tumors. I think these were all Other in the KidsFirst database, but I added the actual diagnoses from our CBTTC database. Maybe this is something @baileyckelly already has solved, though!
What data file(s) does this issue pertain to?
pbta-histologies.tsv
What release are you using?
v15
Put your question or report your issue here.
The
short_histology
label as it currently is used contains a category called "Other", which includes tumor histologies of vastly different types, which can lead to improper grouping if it is used as the main label for a figure or analysis. For example, it includes "Benign tumors" (as labeled inbroad_histology
) such as Fibromas and Dermoid Cysts, but also "Metastatic secondary tumors", which I would imagine we would almost never want to group together.We may want multiple "Other" categories, such as "Other-benign"?
I do not know if this may be resolved as part of #637, but any resolution will almost certainly result in breaking changes, not least for the color palettes in #622.
(tagging in @allisonheath for coordination)
The text was updated successfully, but these errors were encountered: