Back to skills
SKILL.md
End Reason
ASecurityOxford Nanopore read end reason QC analysis. Use when analyzing nanopore sequencing quality, checking for adaptive sampling efficiency, investigating read termination patterns, diagnosing sequencing issues, or running QC on POD5/Fast5 data. Integrates with ont-experiments for provenance tracking via Pattern B orchestration. Supports signal_positive, unblock_mux_change, data_service_unblock_mux_change analysis with quality thresholds.
- 666 stars
- 0 votes
- 0 copies
- 2 views
- Added February 8, 2026
Security analysis
100/100npx -y skills add majiayu000/claude-skill-registry --skill end-reason --agent claude-codeAre you the author of End Reason?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/majiayu000-end-reason)---
name: end-reason
description: Oxford Nanopore read end reason QC analysis. Use when analyzing nanopore sequencing quality, checking for adaptive sampling efficiency, investigating read termination patterns, diagnosing sequencing issues, or running QC on POD5/Fast5 data. Integrates with ont-experiments for provenance tracking via Pattern B orchestration. Supports signal_positive, unblock_mux_change, data_service_unblock_mux_change analysis with quality thresholds.
---
# End Reason Analysis - ONT QC
Analyze Oxford Nanopore read end reasons for quality assessment and adaptive sampling efficiency.
## Integration
Run through ont-experiments for provenance tracking:
```bash
ont_experiments.py run end_reasons exp-abc123 --json results.json --plot qc.png
```
Or standalone:
```bash
python3 end_reason.py /path/to/data --json results.json
```
## End Reason Categories
| End Reason | Description | Expected % |
|------------|-------------|------------|
| `signal_positive` | Normal completion | 80-95% |
| `unblock_mux_change` | Adaptive sampling rejection | 0-15% |
| `data_service_unblock_mux_change` | Basecall-triggered rejection | 0-10% |
| `mux_change` | Pore mux change | 1-5% |
| `signal_negative` | Signal lost | <5% |
## Quality Assessment
| Status | Criteria |
|--------|----------|
| ✓ OK | signal_positive ≥75% |
| ⚠ CHECK | signal_positive <75% or anomalies |
| ✗ FAIL | signal_positive <50% |
## Options
| Option | Description |
|--------|-------------|
| `--json FILE` | Output JSON summary |
| `--csv FILE` | Output per-read CSV |
| `--plot FILE` | Output bar chart (PNG/PDF) |
| `--format FORMAT` | Force format (pod5, fast5, summary) |
| `--quick` | Sample 10k reads only |
## Supported Formats
- POD5 (fastest, recommended)
- Fast5 (legacy)
- sequencing_summary.txt
## Output JSON
```json
{
"total_reads": 15000000,
"quality_status": "OK",
"signal_positive_pct": 90.0,
"unblock_mux_pct": 8.0,
"data_service_pct": 1.0,
"end_reasons": {
"signal_positive": {"count": 13500000, "pct": 90.0},
"unblock_mux_change": {"count": 1200000, "pct": 8.0}
}
}
```
Attribution
Comments
Loading comments…
Newsletter
Get the weekly list of safe new skills, and the ones we pulled.