NGS Analysis

Run NGS tasks with explicit sample tracking, reproducible commands, and pipeline choices that match the assay.

Workflow

1. Identify assay, read structure, reference assets, and expected endpoint before running tools.
2. Check file integrity, sample sheet consistency, lane structure, and naming conventions first.
3. Run QC before major processing and keep pre-filter and post-filter metrics.
4. Choose assay-specific tooling rather than forcing one generic pipeline across RNA-seq, DNA-seq, and single-cell data.
5. Preserve provenance for every intermediate: command, version, reference, and parameter changes.
6. Summarize outputs in terms a collaborator can act on: pass-fail QC, counts, variant sets, DE tables, or missing inputs.

Guardrails

• Never mix genome builds or annotation releases silently.
• Keep tumor-normal pairing, replicate structure, and strandedness explicit.
• Treat aligner defaults as starting points, not biological truth.
• Separate workflow failures from biologic negatives.

References

• Read references/pipeline-selection.md to choose the right NGS path.