@pimentel, a new minor issue that should be addressed before the next major release:
A user on the google group (see the relevant thread) was concerned about missing bootstraps when viewing the summary table for his sleuth object on the shiny App. There were no other error messages.
After working with one of his HDF5 files, I realized that the current procedure in sleuth_prep skips over recording how many bootstraps are present, so that the obj$kal kallisto list will always have an empty/NULL list for the kal$bootstrap. So, this means the summary method (see the relevant line of code) will always return zero.
This is probably easy to fix (always record the number, and make it an attribute of the kallisto object that is directly checked in the summary method), but there is some ambiguity: do we want to always report the number of bootstraps present? Or if the user sets max_bootstrap, do we want to report the actual number used? Or should we report both?
pimentel
commented
6 years ago
@pimentel, a new minor issue that should be addressed before the next major release:
A user on the google group (see the relevant thread) was concerned about missing bootstraps when viewing the summary table for his sleuth object on the shiny App. There were no other error messages.
After working with one of his HDF5 files, I realized that the current procedure in
sleuth_prepskips over recording how many bootstraps are present, so that theobj$kalkallisto list will always have an empty/NULLlist for thekal$bootstrap. So, this means the summary method (see the relevant line of code) will always return zero.This is probably easy to fix (always record the number, and make it an attribute of the kallisto object that is directly checked in the summary method), but there is some ambiguity: do we want to always report the number of bootstraps present? Or if the user sets
max_bootstrap, do we want to report the actual number used? Or should we report both?