Skip to content

Pipelines

iFlow runs two kinds of pipelines against your samples:

  • Secondary pipelines take raw sequencer output — FASTQ or aligned BAM — and produce called variants (VCF), per-sample QC metrics, and coverage data. This is the expensive step: alignment, variant calling, joint genotyping.
  • Tertiary pipelines take VCF (plus optional metadata) and produce annotated, classified, report-ready variants. Each clinical reporter — hereditary, somatic, myeloid, PGx — is its own tertiary pipeline.

Why the split matters

The split lets you re-run just the tertiary step against the same secondary outputs. New annotation version? Updated ACMG criteria? Improved evidence database? Re-run tertiary, keep the (expensive) secondary results from disk. iFlow tracks this dependency for you — tertiary runs always reference the secondary run they consume.

See also