🧪 Requirement: Test Data for External Validation
Summary
At present, the protEncode pipeline does not include any small-scale example dataset for testing and validation by other users.
This limits the ability for contributors and new users to run the package end-to-end and verify functionality without access to the full internal data.
Proposed Solution
- Create a minimal test dataset (e.g. 3–5 protein sequences, 2–3 variant files, and corresponding metadata).
- Store it under
data/test/ or make it accessible via a small GitHub release or Zenodo link.
- Add clear documentation in the
README.md describing how to run protEncode using this test data.
Goals
- ✅ Enable reproducibility and automated CI testing.
- 🚀 Allow new contributors to confirm environment setup and expected outputs quickly.
- 📊 Provide a consistent benchmark for future development and regression testing.
Suggested Contents for Test Data
| File Type |
Description |
| FASTA |
A few short protein sequences (e.g. 3–5) |
| Mutation file |
Example input in the expected format |
| Metadata table |
Includes gene, UniProt ID, and mutation position |
| Example output |
Mock embeddings or summary file for validation |
Priority
Medium
Status
Open for discussion
🧪 Requirement: Test Data for External Validation
Summary
At present, the
protEncodepipeline does not include any small-scale example dataset for testing and validation by other users.This limits the ability for contributors and new users to run the package end-to-end and verify functionality without access to the full internal data.
Proposed Solution
data/test/or make it accessible via a small GitHub release or Zenodo link.README.mddescribing how to runprotEncodeusing this test data.Goals
Suggested Contents for Test Data
Priority
Medium
Status
Open for discussion