- lineplot, boxplot:
- Use dv.manager's switch2mod instead of deprecated switch2.
- lineplot, boxplot:
- Make receiver_id accept module identifiers instead of labels.
- lineplot:
- Prevent spurious reactive update.
- lineplot:
- Per-column default visit values. Breaking change, as it changes the expected type of
default_visit_val
. - Fixes order of x-axis visit ticks for continuous variables.
- Per-column default visit values. Breaking change, as it changes the expected type of
- lineplot hotfix:
- Prevent suspendWhenHidden of dynamic UI.
- lineplot:
- Y axis log projection menu option.
- New
cdisc_visit_vars
parameter that accounts for missing Study Day 0 in CDISC datasets. - New
default_transparency_value
parameter. - Remove support for data dispatchers in favor of dataset names.
- WFPHM:
- Fixes the error in conditional panels that prevented conditional panels in other modules to work properly.
- Lineplot:
- Fix overlap of plot and data listings.
-
Lineplot:
- Prevent false-positive opaque error message at module startup.
-
Correlation heatmap:
- Provide clear error message when handed records with identical subject IDs, category, parameter and visit values.
- First Github release
- Renamed to dv.explorer.parameter
-
Correlation heatmap:
- NA and Inf values are removed before the scatterplot to avoid spurious points
-
Lineplot:
- Lineplot does no longer crash when dataset is empty
- Populates inline titles with column names when labels are not available
- Brush selection behaves correctly when sidebar is expanded in dv.manager and it does not crash when the selection is empty
- Prevents visit reset when dataset changes in module manager
-
Migrated Waterfall plus heatmap module to dv.biomarker.general
-
Migrated dv.roc to dv.biomarker.general
-
Lineplot:
- Added summary data listing with aggregated centrality (mean, median, ...) and dispersion (standard deviation, confidence interval, ...) scores.
-
Correlation heatmap:
- Selection interface now allows multiple, independent cross-visit and cross-parameter combinations.
- Faster plot rendering.
- Cleaner graphic output and better management of long axis labels.
-
Common:
- Each module now explicitly specifies the value column used to generate the figures.
- Default values are included in all modules.
- Menus have been redesigned using shinyvalidate to provide more informative feedback regarding errors or missing selections.
-
Boxplot:
- Added functionality for saving/loading visualization states.
-
Lineplot:
- Added functionality for integration with other modules
- Supports custom centrality and dispersion measures
-
Fix:
- Rectified issue where Boxplot added jitter in the y-axis unnecessarily.
- Renamed
value_var
tovalue_vars
in the Scatter Plot Matrix module to adhere to DV standards. - Improved error handling for correlation calculations in Scatter plot matrix
-
Other:
- Replaced the magrittr pipeline
%>%
with the base pipeline|>
. - Resolved the issue with
subset_ds
for single column selections. - Implemented new synthetic datasets in mock_* functions and testing procedures.
- Replaced
dv.selector
with locally developed selectors
- Replaced the magrittr pipeline
- Initial release with the following modules:
- Boxplot:
vignette("boxplot")
- Scatterplot:
vignette("scatterplot")
- Lineplot:
vignette("lineplot")
- Scatterplotmatrix:
vignette("scatterplotmatrix")
- Forest plot:
vignette("forest")
- Scatterplot heatmap plot:
vignette("correlation_heatmap")
- Boxplot: