The QC process is executed using the FAIRlyz QC toolkit which is installed locally in the compute environment where the data resides and is ready to be analyzed.
🛠️ Before You Begin: Install FAIRlyz QC
Install the FAIRlyz QC Tool on your computer. The tool runs as a Docker application — ask your IT team to help with the installation if needed.
🔒 What is Data Visiting QC?
FAIRlyz QC validates your research data files directly on your computer — nothing is uploaded or copied to any server. The process is fast, private, and designed for NIH dbGaP submission standards.
⚡ The Four Core Steps
When your dictionary and mapping files are ready, the entire workflow takes only four steps.
Below is a screenshot of FAIRlyz QC at the start of your QC process.
1. 📁 Select Your Files
Drag and drop your three files into the app:
- 📊 Data file — your research dataset (.csv)
- 📖 Dictionary file — describes each variable in your data
- 🗺️ Mapping file — connects your variables to standardized ontology terms
The app auto-detects file types by filename. No configuration required.
2. ✅ Run Dictionary QC
Click Dictionary QC. The app validates your dictionary against NIH dbGaP standards in seconds. A QC Score shows how well your dictionary meets the submission requirements.
3. ✅ Run Mapping QC
Once Dictionary QC passes, click Mapping QC. The app checks that your variables are correctly mapped to recognized ontology terms. Your combined QC score updates automatically.
4. 🔄 FAIRlyz Sync
When both QC scores meet the threshold, click FAIRlyz Sync to push your validated results to the database. Your raw data files never leave your computer — only QC results and metadata are synchronized.
🛠️ Optional: Generate or Edit Files
Don’t have a dictionary or mapping file yet? The app generates them for you:
- 📖 Dictionary — auto-generated from your data file using local rule-based analysis. No AI, nothing leaves your machine.
- 🗺️ Mapping — generated by AI that reads your dictionary only (never your raw data). Built-in editors let you review and refine both files before running QC.
🔒 Why Data Visiting?
| Traditional Approach | FAIRlyz Data Visiting |
|---|---|
| Files uploaded to a server | Files read from your local disk |
| Data leaves your organization | Data stays on your computer |
| Privacy and compliance risk | Maximum privacy by design |
| Your data may be rejected after sharing | Provide a QC report before sharing data |
Only QC scores, statistics, and metadata you choose to share are ever shared in a report.
Dictionary QC results — your dictionary is validated against NIH dbGaP standards.
Mapping QC results — your variables are semantically validated against standardized ontology terms.
🏅 Standards Compliance
FAIRlyz validates against:
- ✅ NIH dbGaP submission standards
- ✅ FAIR principles — Findable, Accessible, Interoperable, Reusable
- ✅ Ontology standards — standardized medical and scientific terminology