Skip to contents

Changelog

Source: NEWS.md

nprcgenekeepr 1.1.0.9000 (20260126)

  • Shiny application
    • New Potential Parents tab. For a loaded pedigree it identifies in-colony animals (fromCenter) that have at least one unknown parent and lists candidate sires and dams, screened by estimated conception date (birth minus the maximum gestational period). A numeric Maximum Gestational Period (days) input (default 210, the rhesus upper bound) is overridable per run; a Find Potential Parents button computes on the current pedigree; results display in a sortable table and download as CSV. The tab degrades gracefully when no pedigree is loaded, when the pedigree lacks the fromCenter colony-origin field, or when no animal has an unknown parent. This wires the exported getPotentialParents() (including the gestation-derived dam-exclusion window) into the app. (#48)
    • New ORIP Reporting tab. Provides formatted colony summaries for submission to the NIH Office of Research Infrastructure Programs (ORIP): site information, a colony summary table (totals by sex with founder counts via isFounder()), and genetic-diversity metrics (mean kinship and mean genome uniqueness) computed from the loaded pedigree and genetic-value results. Two Export buttons download the ORIP report and the demographics as CSV. This mounts the previously unwired modORIPReporting module pair into the application. The tab is ONPRC-specific: it is shown only when an actual site configuration file identifies the colony as ONPRC, and is hidden at other sites and when no configuration file is present. (#47, #49)
    • Fixed a startup crash. The application no longer fails to launch when a site configuration file written in the documented format (comment lines, blank lines, and multi-line / quoted / comma-separated values, as in example_nprcgenekeepr_config) is present in the user’s home directory. Configuration is now read through the same tolerant parser used by getSiteInfo(), via the new loadSiteConfig(), which logs a warning and leaves the configuration unset (rather than aborting boot) if the file is missing or malformed. The previous loader used read.table(sep = "="), which assumed a strict two-column table and stopped with “line N did not have 2 elements”. (#50)
    • The About panel now shows the installed package version dynamically (via getVersion()), reading it from the package DESCRIPTION instead of a hard-coded string. The previous static “Version 1.0.8” had drifted out of date; deriving it at run time keeps it from going stale again.
  • Data input / quality control
    • IDs may no longer contain a period (“.”). qcStudbook() and geneDrop() now reject id/sire/dam values that contain a period; with reportErrors = TRUE, the offending values are reported in errorLst$invalidIdChars. Periods cause problems across software environments (R column-name/formula parsing, file extensions, namespaces, regular expressions). All automatically generated IDs remain period-free. (NEW-45)
    • geneDrop() now rejects duplicate animal IDs with a clear error. Animal IDs uniquely identify animals (already enforced upstream by qcStudbook() via removeDuplicates() and by kinship()); a duplicate id previously triggered the cryptic base-R error “duplicate ‘row.names’ are not allowed”. (NEW-46)
    • Reading an animal list or pedigree file whose final line has no trailing newline no longer emits the confusing “incomplete final line found by readTableHeader” warning. Every row, including the last, was always read correctly – the warning was harmless noise. getPedigree(), getGenotypes(), getFocalAnimalPed(), and the Shiny file upload now suppress only that one warning while letting every other read warning through. (#4)
  • Genetic value analysis
    • summarizeKinshipValues() now reports the secondQuartile column as the lower hinge (fivenum()[2], approximately the first quartile) instead of silently duplicating min (fivenum()[1]). The thirdQuartile column (the upper hinge) was already correct. (NEW-16)
    • calcFE(), calcFG(), and calcFEFG() no longer duplicate the founder- contribution algorithm and the partial-parentage guard they shared verbatim; both now live in a single internal helper. Results, signatures, and error messages are unchanged. (NEW-13 / NEW-23)
  • Pedigree curation
    • Added two exported helpers, isFounder() and getFounders(), that identify pedigree founders (animals whose sire and dam are both unknown). isFounder(ped) returns the logical mask and getFounders(ped) returns the founder id values. The founder predicate had been written inline in a dozen places; it is now defined once and reused throughout the genetic-value and reporting functions. (PED-1 / NEW-17)
    • The Pedigree Browser’s “Trim pedigree based on focal animals” option now includes both the ancestors and the descendants of the focal animals (it previously included ancestors only). A new exported helper, getDescendantPedigree(), returns the transitive descendants of a set of probands – the downward mirror of getProbandPedigree(). Trimming is strict-lineal: collateral relatives (siblings, cousins, mates) are not added. (NEW-47)
    • The format of the auto-generated placeholder IDs created for unknown parents (see addUIds()) is now configurable from a single source of truth shared by ID generation and detection. Two new exported helpers, getAutoIdFormat() and setAutoIdFormat(), read and set the sprintf format (default "U%04d"); with no configuration all existing behavior is unchanged. Detection is now centralized in one internal predicate used by removeAutoGenIds(), the Pedigree Browser display filter, reportGV() founder counts, and obfuscateId() – replacing eight scattered string literals and reconciling their formerly inconsistent case-handling to case-sensitive (matching the uppercase prefix that generation emits). (NEW-48 / issue #44 / #38)
    • getPotentialParents() now selects candidate dams using a gestation-derived exclusion window driven by its existing maxGestationalPeriod parameter, replacing a fixed half-year window. A female who delivered another offspring within maxGestationalPeriod days of a focal animal’s birth is excluded as a candidate dam (a female bears one offspring at a time), so dam selection now responds to the species’ gestation length rather than a hard-coded +/- 182.5 days. Sire selection already used this parameter; the sire (presence at conception) / dam (presence at birth) exit-check asymmetry is intentional and now documented. (NEW-49 / issue #31)
  • Major changes
    • Architectural Changes
      • Modular Shiny Architecture
        • Refactored monolithic Shiny application into discrete, testable modules using shiny::moduleServer()
        • New module files:
          • modInput.R - Data input and QC processing
          • modPedigree.R - Pedigree browser with trim/filter capabilities
          • modPyramid.R - Age-sex pyramid visualization
          • modGeneticValue.R - Genetic value analysis (mean kinship, genome uniqueness)
          • modSummaryStats.R - Summary statistics with interactive visualizations
          • modBreedingGroups.R - Breeding group formation using groupAddAssign()
          • modORIPReporting.R - ORIP reporting module
        • New appServer.R and appUI.R orchestrate module communication via shared reactive values
        • runModularApp() provides entry point for modular application
        • Monolith retired (Phase 9). The legacy monolithic application (inst/application/) has been deleted. runGeneKeepR() is now a soft-deprecated alias that launches the modular application via runModularApp(); existing zero-argument calls continue to work. Also removed the now-unused exports getLogo(), shouldShowErrorTab(), modMinimalTestUI(), and modMinimalTestServer(), plus the unexported getMinParentAge(). (XARCH-1 / issue #27)
    • New Features
      • Dynamic Tab Management
        • Error List and Changed Columns tabs appear/disappear dynamically based on QC results
        • Uses insertTab()/removeTab() for cleaner UI when no errors present
      • Enhanced QC Pipeline
        • runQcStudbook() wrapper provides UI-friendly error reporting
        • processQcStudbookResult() transforms QC output for display
        • shouldShowChangedColsTab() helper function
      • Improved Visualizations
        • getBoxWhiskerDescription() provides educational popover content for box plots
        • savePlotToFile() supports PNG, PDF, and SVG export
        • Enhanced pyramid plots with getPyramidPlot()
      • Genetic Value Analysis tab parity (modular app)
        • Exposed the genome-uniqueness threshold as a user control with a default of 4, matching the legacy application. The modular app previously hard-coded a threshold of 1, so default genome-uniqueness values from runModularApp() now match the legacy app. (The analytical reportGV() default, guThresh = 1, is unchanged.)
        • Added a subset filter (view by animal IDs) and an “Export Subset” download, matching the legacy “Filter View” / “Export Current Subset”.
        • Changed the default gene-drop iterations to 1000 for legacy parity (was 5000); removed an inert “Minimum breeding age” slider that had no effect on the analysis.
      • Utility Functions
        • safeExecute() - Error-handling wrapper for module operations
        • logModuleEvent() - Structured logging with futile.logger integration
        • makeFounderStatsTable(), makeGeneticSummaryTable() - Table generators
    • Testing Improvements
      • Added shiny::testServer() unit tests
      • ~145 new/modified test files with comprehensive module coverage
      • Tests for edge cases: NULL inputs, empty pedigrees, single-animal scenarios
      • Strict TDD development process used for all new features
  • Minor changes
    • New Dependencies
      • Added to Imports: bslib, DT, ggplot2
      • Added to Suggests: shinytest2
    • Bug Fixes
      • Fixed undefined global variables in ggplot2 aes() calls
      • Fixed column name expectations in genetic value tests (meanKinship/genomeUniqueness)
      • Network-dependent tests now skip gracefully
    • Documentation

nprcgenekeepr 1.0.8 (20250723)

CRAN release: 2025-07-26

  • Minor changes
    • Added returned value descriptions for all functions within R directory where formerly missing.
    • Changed unit test for get_elapsed_time_str() to use a mocked version of proc.time()

nprcgenekeepr 1.0.7 (20250506)

CRAN release: 2025-04-24

  • Minor changes
    • Added returned value descriptions for all functions where formerly missing.
    • Removed extraneous spaces from DESCRIPTION file.
    • Exposed all examples in roxygen2 comments by removing and and . The example with runGeneKeepR() is protected with if (interactive()) {}.

nprcgenekeepr 1.0.6 (20241215)

  • Minor changes
    • Update version in preparation for CRAN submission
    • Added article demonstrating Simulated Kinships with Partial Parentage
    • Added use of CICD pipeline as GitHub Actions
      • lintr pipeline
      • R CMD check pipeline with multiple R environments and versions
      • pkgdown pipeline
    • Added several unit tests
    • Cleaned up code based on lintr feedback
    • Added example deidentified pedigree data 2022-05-02_Deidentified_Pedigree.xlsx, 2022-05-02_Deidentified_Pedigree_focal_animals.csv, deidentified_jmac_ped.csv (text, except for dates, are in double quotes), deidentified_jmac_ped_edited.csv (edited to remove double quotes).
    • Made getVersion() more robust.
    • Abstracted out removal of auto generated Ids in preparation of allowing the user to define how auto generated Ids will be formed.
    • Added some quality assurance badges to README.
    • Added CRAN status badge to README.
    • Stopped using travis-ci and started using GitHub Actions with Rhub.yaml file for checking on Rhub.

nprcgenekeepr 1.0.5.9004 (20221213)

  • Minor changes
    • Changed method used to test class of object to use inherits().
    • Corrected getPedDirectRelative() so that all direct relatives are found. Supplemented unit tests for more direct relative types.
    • Added unit tests for trimPedigree().
    • Changed call as.character(date_object) to format(date_object) in getDatedFileName.R to prepare for newer code in development version of
    • Technical edits of R code based on lintr::lint_dir(“R”)

nprcgenekeepr 1.0.5.9003 (20220625)

  • Minor changes
    • Removed dependency on gdata.
    • Removed getMinParentAge() as it was never used.
    • Starting to replace rbind() with rbindlist() from data.table were possible.

nprcgenekeepr 1.0.5.9002 (20220425)

  • Minor changes
    • Added use of data.table in an effort to reduce memory use and CPU use for estimation of kinship values.
    • Functions were refactored and the ability to handle larger simulations resulted.

nprcgenekeepr 1.0.5.9001 (20210830)

  • Major changes
    • Added ability to use simulation to estimate the kinship values of animals with incomplete parental information that are known to have been born within the colony. These animals may have 0 or 1 known parents but have a value in the pedigree file or database for the fromcenter or fromCenter field of “Y”, “YES”, “T”, or “TRUE”.
  • Minor changes
    • Increase unit test coverage primarily to include more rare events and events that should not happen and are trapped and result in errors.
    • Changed to travis-ci.com

nprcgenekeepr 1.0.5 (20210328)

CRAN release: 2021-03-31

  • Major changes – none
  • Minor changes
    • CRAN submission primarily in response to a change in shiny 1.6 that removed an internal shiny function (shiny:::%OR%) and replaced it with rlang::%||%
    • Stale URL in historical documentation that were causing notes to be generated in automated tests have been removed.
    • A URL referring to Terry Therneau’s page was updated from “http” to “https”.
    • I have incremented the version from 1.0.4 (github.com only version) to 1.0.5, updated NEWS to reflect the changes, and updated all documentation to reflect the version change.

nprcgenekeepr 1.0.4.9003 (20210318)

  • Major changes – none
  • Minor changes
    • Testing .travis.yml code change to get textshaping to build on all systems..
    • Cleaned up .travis.yml in response to syntax checking on travis.org.
    • Added markdown to suggest due to new changes in knitr.

nprcgenekeepr 1.0.4 (20210318)

  • Major changes – none
  • Minor changes
    • Added suppression of warnings from DT at beginning of server.R since it is unlikely for anyone to call affected functions in the controlled environment.
    • Changed call to shiny:::%OR% to rlang::%||% in server.R since the update to 1.6 of shiny broke the code. Thanks to Dan Metzger of Wisconsin National Primate Research Center.

nprcgenekeepr 1.0.3 (20200526)

CRAN release: 2020-06-02

  • Major changes – none
  • Minor changes
    • CRAN re-submission: responded to the two requests provided by reviewer
      • I have removed the capitalization from “Genetic Tools for Colony Management” and “Genetic Value Analysis Reports” within DESCRIPTION.
      • I have removed the conditional installation of DT from the ui.R file.
    • I have incremented the version from 1.0.2 to 1.0.3, updated NEWS to reflect the changes, and updated all documentation to reflect the version change.

nprcgenekeepr 1.0.2 (20200517)

  • Major changes – none
  • Minor changes
    • CRAN re-submission: responded to all requests provided by reviewer
      • I have not changed the capitalization of Shiny in the description section of the DESCRIPTION file as it is the name of the type of application and is not being used as the name of the package. The use of the capitalization is consistent with the capitalization used within the documentation for the shiny package (?shiny, See the Details section, first sentence where it is used as the type of tutorial.) and all documentation and tutorials provided by the author and RStudio where it is capitalized everywhere except when referring to the package.
      • I have continued to use dontrun for the following examples:
        • runGeneKeepr(), which starts the Shiny application
        • getFocalAnimalPed(), which is dependent on a valid LabKey instance, a proper configuration file, and a .netrc or _netrc authentication file.
      • I have exchanged dontrun for donttest for the following examples:

nprcgenekeepr 1.0.1 (20200510)

  • Major changes – none
  • Minor changes
    • CRAN re-submission: responded to all requests provided by reviewer
      • Reduced the time required for unit test from over 12 minutes to 21.6 seconds by skipping those test dependent on stochastic creation of simulated pedigrees and breeding groups when not running on my system.
      • Reduced the time to run examples and create vignettes by reducing the number of stochastic modeling iterations by orders of magnitude without reducing the examples provided for user-facing functions.
      • Checking (–as-cran –run-donttest) Duration: 2m 21.8s on my system.
      • The files with the Rd-tag of \arguments missing do not take arguments.
      • Corrected private referencing (:::) for exported functions.
      • Exported all functions used in examples to remove private referencing (:::).
      • Removed all single quotes on names, abbreviations, initialisms, and, acronyms.
      • The phrase Electronic Health Records (EHR) is the name of a module within LabKey, which this software can use as a source of pedigree information so the capitalization is appropriate.
      • Two exported functions used by server.R to call tabpanel() do not have examples.

nprcgenekeepr 1.0 (20200415)

  • Major changes – none
  • Minor changes
    • CRAN submission

nprcgenekeepr 0.5.43 (20200414)

  • Major changes – none
  • Minor changes
    • Final preparation for CRAN submission

nprcgenekeepr 0.5.42.9012 (20200412)

  • Major changes – none
  • Minor changes
    • Updated unit test for dataframe2string to account for change in age of a sire from 8.67 to 8.66 years.
    • Renamed tutorials.

nprcgenekeepr 0.5.42.9011 (20200409)

  • Major changes – none
  • Minor changes
    • Build failed on Travis-ci due to unit test failure but the test has never failed and does not fail on other builds. Removed set_seed() to see if that helps.
    • Fixed GitHub issue 3
    • Added additional explanatory text from Matt Schultz edits for the Colony Manager version of the Shiny tutorial.

nprcgenekeepr 0.5.42.9010 (20200405)

  • Major changes – none
  • Minor changes
    • Added code to address issue 1 (GitHub). See comment 1 for details, but more should be done.
    • Refreshed Shiny_app_use.Rmd to reflect changes since November 2019.

nprcgenekeepr 0.5.42.9009 (20200402)

  • Major changes – none
  • Minor changes
    • Wrapped example for makeExamplePedigreeFile with \dontrun{} because R 4.0.0 alpha was leaving the side effect of the dataframe stored in a CSV file named as the text of the next line.

nprcgenekeepr 0.5.42.9008 (20200321)

  • Major changes – none
  • Minor changes
    • Changed dependency to R >= 3.6 since caTools is not available for R < 3.6.

nprcgenekeepr 0.5.42.9007 (20200319)

  • Major changes – none
  • Minor changes
    • Changed warnings unit test for getLkDirectAncestors to work with Windows.

nprcgenekeepr 0.5.42.9006 (20200319)

  • Major changes – none
  • Minor changes
    • Completed examples in function documentation
    • Corrected spelling of several word throughout found with spelling::spell_check_package(".").

nprcgenekeepr 0.5.42.9005 (20200201)

  • Major changes – none
  • Minor changes
    • Added examples to function documentation
    • Added ColonyManagerTutorial.Rmd initial draft, which is copy of shiny_app_use.Rmd. It is to be converted for use by colony managers.

nprcgenekeepr 0.5.42.9004 (20200201)

  • Major changes – none
  • Minor changes
    • Added examples to function documentation
    • Added obfuscated rhesus pedigree and rhesus haplotypes to use in examples

nprcgenekeepr 0.5.42.9003

  • Major changes – none
  • Minor changes
    • Renamed local and remote repositories from nprcmanager to nprcgenekeepr.

nprcgenekeepr 0.5.42.9002

  • Major changes
    • Changed name of package to nprcgenekeepr. This required changing of many of the supporting files and functions. Having good unit test coverage of the functions (739 test with > 90 percent coverage) made this possible.
    • This is the last version under the nprcmanager repository name.
    • Conversion worked
      • Running the build check had OK: 739; Failed: 0; Warnings: 0; Skipped: 0
  • Minor changes – none