Approaches to Assay Processing Package Validation
summary
Five concrete steps for validating internal R packages at SCHARP: write numbered specifications, annotate functions with ownership metadata, author human-readable test cases, have a third party implement tests, and compile results into a single validation artifact. Workflow uses roxygen2 for ownership tags, testthat for executable tests and reporter objects, and rmarkdown to generate a signed PDF that maps specifications to pass/fail outcomes, supports FDA 21 CFR 11 style needs, and is used on live trial data.
AI-written summary · about
chapters (7)
- 0:00Start
- 4:53Specifications What are the requirements of the package?
- 5:24Recording specifications Record Specifications in markdown
- 7:03Documenting code ownership
- 9:33Documenting Test Cases
- 10:12Modular Test Case Documentation
- 10:32Test Coding Implement Test Cases
related talks
18mR Package Assessment: Lessons from Pharma
05mCoding Tools for Industry R&D – Development Lessons from an Analytical Lab
01h 32mCreating Production-Ready Shiny Apps
15mPackages and Process
17mBreaking Barriers: Adopting R in Biotech with Posit
18mHow non-programmers can make sparks fly when using testthat during package development
19mElevating Public Health Decision-Making with R Packages
46mDevelopment and Testing Strategies