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
Goal: The goal of this issue is to create unit tests for high-throughput transcriptomics (HTTR) signature data such that we ensure that as the codebase of tcplfit2 is updated that the updates do not break/alter the results previously obtained in previous analyses with tcplfit2.
Needs for finishing this ticket:
Choose a manageable subset of the overall data provided by the HTTR team - verify the chosen set covers many (if not all) scenarios of interest
Create an RData file for the package (may want to ensure we do NOT want this to be exported from the package)
Create a unit test ensuring that the results for HTTR remain consistent from version to version of tcplfit2
The text was updated successfully, but these errors were encountered:
utilize the previous used criteria for HTPP to fulfill needs we need to meet
if there are any signature categories that can be identified then do random selections in a stratified manner - Grace mentioned there were GO and CMAP ( choose 10 genes per category, then 30 other randomly selected genes)
like with the genes - if a baseline of 50 signatures (covering the previous bullet) is relatively fast, then may look to bump up to 100 signatures (20 per category and 60 other random signatures).
Goal: The goal of this issue is to create unit tests for high-throughput transcriptomics (HTTR) signature data such that we ensure that as the codebase of tcplfit2 is updated that the updates do not break/alter the results previously obtained in previous analyses with tcplfit2.
Needs for finishing this ticket:
The text was updated successfully, but these errors were encountered: