agentsclimarketplace

Integration test development

Skill HolobiomicsLab/asb-skill-collections/packs/metabolomics/ms-generic/skills/integration-test-development

Curated, evidence-grounded skill and software-tool collections for scientific AI agents, generated by the AgenticScienceBuilder

Install
npx -y skills add HolobiomicsLab/asb-skill-collections --skill integration-test-development

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 14 stars14 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when when you have implemented or modified a data ingestion module (e.g., mzML parser) and need to verify that file deserialization produces correct internal representations.

The file declares its own license as CC-BY-4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

8.2 KB, as published. Nobody here has run it

integration-test-development

Summary

Develop and execute integration tests to validate that parsed mass spectrometry raw data files (mzML format) correctly map metadata, spectral arrays, and quantitative fields into a software's internal data structures. This skill ensures reproducibility and correctness of file format ingestion pipelines across build configurations and CI/CD environments.

When to use

When you have implemented or modified a data ingestion module (e.g., mzML parser) and need to verify that file deserialization produces correct internal representations. Specifically, when peak counts, mass accuracy, retention time values, instrument metadata, m/z arrays, and intensity arrays must match reference expectations from the source file without vendor-specific SDK dependencies.

When NOT to use

  • Input is proprietary vendor format (ABF, RAW, D folder) — unit testing for vendor SDKs is not currently set up; use only for mzML, CDF, or ABF formats supported in vendor unsupported mode.
  • Partial or stub functionalities without full parser implementation — unit testing infrastructure for partial functionalities is not currently set up for trial.
  • Testing requires reproducible builds across configurations other than version 5 series — only the version 5 series of MS-DIAL has reproducible builds that can be guaranteed openly.

Inputs

  • mzML-format raw data file (reference file with known metadata and spectral characteristics)
  • Implemented mzML parser module (source code or compiled library)
  • Build environment configuration (.NET Framework 4.7.2 or .NET 6, Visual Studio or VSCode)

Outputs

  • Integration test suite (unit tests targeting parser output)
  • Test execution report (pass/fail per assertion, coverage summary)
  • CI/CD build log with test results from GitHub Actions
  • Validated internal data structure instances (parsed mzML objects)

How to apply

Load a reference mzML file using the implemented parser and extract parsed metadata (instrument type, acquisition parameters) and spectral arrays (m/z values, intensity values). Write test assertions that compare parsed peak counts, mass accuracy (typically ±5 ppm tolerance for Orbitrap or similar instruments), and retention time fields against expected ground-truth values from the reference file. Group tests by data category (metadata, peak properties, array consistency) and execute them via the CI/CD pipeline (GitHub Actions) against the version 5 series build configuration to ensure reproducibility. Verify that tests pass on both Debug vendor unsupported and Release configurations where applicable.

Related tools

  • GitHub Actions (CI/CD pipeline execution environment for running integration tests via the dotnet_test.yml workflow) — https://github.com/systemsomicslab/MsdialWorkbench
  • .NET Framework 4.7.2 / .NET 6 (Runtime and language framework for compiling and executing parser and test code)
  • Visual Studio 2022 (IDE for writing, debugging, and locally executing integration tests before CI/CD submission)
  • MS-DIAL MsdialWorkbench repository (Source code repository containing mzML parser implementation, test frameworks, and build recipes) — https://github.com/systemsomicslab/MsdialWorkbench

Evaluation signals

  • All test assertions pass: parsed peak count equals reference count (exact match), mass accuracy falls within tolerance (e.g., ±5 ppm for reference m/z values), retention time fields match reference to within ±0.1 minute.
  • Metadata extraction correctness: instrument type, acquisition parameters, and scan-level metadata are correctly deserialized and stored in internal structures without null or default-value artifacts.
  • Spectral array consistency: m/z array and intensity array lengths match, m/z values are monotonically increasing or in expected order, intensity values are non-negative and finite.
  • CI/CD reproducibility: test suite executes successfully on GitHub Actions runner under version 5 series configuration; build log shows no warnings related to parser deserialization.
  • No vendor SDK dependencies: tests pass using only mzML format input and do not invoke proprietary vendor libraries or require runtime flags for vendor format support.

Limitations

  • Unit testing infrastructure for partial functionalities is not currently set up for trial; tests must target complete, functional parser implementations.
  • Public version (vendor unsupported) cannot validate against proprietary raw data formats (ABF, RAW, D folders); tests are limited to mzML, CDF, and Reifycs ABF formats.
  • Only version 5 series of MS-DIAL has reproducible builds that can be guaranteed openly; tests should target this version to ensure reproducibility across contributors.
  • Mass accuracy thresholds and peak-matching tolerances are not explicitly specified in the documentation; practitioners must define these based on their instrument specifications and may need empirical calibration.
  • Integration tests verify parser correctness but do not validate downstream metabolomics or lipidomics analysis steps; they focus solely on data ingestion and internal structure population.

Evidence

  • [other] Write integration tests to check that parsed peak counts, mass accuracy, and retention time fields match expected values from the reference mzML file.: "Write integration tests to check that parsed peak counts, mass accuracy, and retention time fields match expected values from the reference mzML file."
  • [other] Load a sample mzML file using the implemented parser and verify that metadata (instrument type, acquisition parameters) and spectral arrays (m/z values, intensity values) are correctly mapped to the internal structures.: "Load a sample mzML file using the implemented parser and verify that metadata (instrument type, acquisition parameters) and spectral arrays (m/z values, intensity values) are correctly mapped to the"
  • [other] Execute tests via GitHub Actions CI/CD pipeline to ensure reproducibility on the version 5 series build.: "Execute tests via GitHub Actions CI/CD pipeline to ensure reproducibility on the version 5 series build."
  • [readme] only the version 5 series of MS-DIAL has reproducible builds that can be guaranteed openly: "only the version 5 series of MS-DIAL has reproducible builds that can be guaranteed openly"
  • [readme] The version we have made public does not utilize the MS vendor's SDK. Please be aware that it only supports input in the mzML format for raw data.: "The version we have made public does not utilize the MS vendor's SDK. Please be aware that it only supports input in the mzML format for raw data."
  • [readme] we apologize that the unit testing aspect for partial functionalities is not currently set up for trial: "we apologize that the unit testing aspect for partial functionalities is not currently set up for trial"

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.