From extracted DNA to a called variant
The consumables that carry a hereditary-cancer sample across the bench — extraction chemistry, library preparation, and the flow cell itself. Educational explainers for the curious; not laboratory protocols, and never medical advice.
Swab to sequencer: the bench workflow
Every hereditary-cancer result begins as a tube of cells and ends as a base-called read. Four consumable-heavy stages sit between. Each links to the products and the science behind it.
- 01
1. Nucleic-acid extraction
Lysis, binding, wash, and elution recover genomic DNA from saliva, buccal cells, or blood. Magnetic-bead systems (e.g. Promega Maxwell, MagPurix) yield clean, high-molecular-weight DNA quantified by fluorometry (Qubit) before anything downstream proceeds.
- 02
2. Library preparation
DNA is fragmented (sonication or tagmentation into ~50–500 bp pieces), end-repaired, and ligated to sequencing adapters. For panels, biotinylated hybrid-capture probes enrich only the genes of interest —
BRCA1,BRCA2,PALB2,TP53and the rest of the panel. - 03
3. Quantification & pooling
Libraries are normalised by qPCR or fluorometric assay so each sample contributes evenly to the pool. Accurate molarity here is what protects per-sample read depth later — under-loaded libraries fall below the 30× floor.
- 04
4. Sequencing
The pooled library is loaded onto a flow cell. Clusters are amplified in situ and sequenced by synthesis; raw reads are then aligned and variants called against a reference, ready for ACMG classification.
Reading depth by application
How much coverage a region needs depends on what you are asking of it. Germline questions tolerate less depth than somatic ones because the expected variant fraction is higher and more predictable.
| Application | Typical depth | Why | Source |
|---|---|---|---|
Germline panel (e.g. BRCA1/BRCA2) | ≥30× | Heterozygous variants sit near a 50% allele fraction and are reliably called at modest depth | PMC / Scientific Reports |
| Whole-genome germline | ~30–35× | Standard for germline SNV/indel/CNV detection across the genome | Manufacturer guidance |
| Somatic / tumour variants | ~100×+ | Low-fraction mosaic and subclonal variants demand far deeper reads | Panel literature |
| Whole-exome (comparison) | ~94% bases ≥30× | Broader target, less uniform capture than a focused panel | PMC7189534 |
Table 1. Indicative per-base coverage targets reported in the panel-sequencing literature.
NovaSeq output, illustrated
Flow-cell capacity has scaled by an order of magnitude across one product generation. More reads per run means more samples multiplexed without sacrificing the per-sample depth floor. Values are approximate single-read maxima from manufacturer specifications.
up to ~2.5B paired-end reads per lane; ~20B per run
smaller-batch runs
mid throughput
26–35B single reads per flow cell
The four consumable families
Each stage of the bench workflow leans on a distinct class of consumable. These are the explainer spokes for this hub.
PCR plates & reagents
Master mixes (polymerase, dNTPs, MgCl₂, buffer), 96- and 384-well plates, and the qPCR chemistry that quantifies libraries before pooling.
ReadPipettes & tips
Where contamination and pipetting error are designed out — filter tips, calibration, and why volumetric accuracy at low microlitres protects read depth.
ReadDNA extraction kits
Magnetic-bead and column chemistries that recover clean, high-molecular-weight genomic DNA from saliva, swab, or blood — the input everything downstream depends on.
ReadSequencing flow cells
The patterned surface where clusters form and bases are read. Flow-cell choice sets the read budget — and therefore how deeply each sample can be covered.
ReadBench glossary
A few terms that recur across the four spokes.
- Tagmentation
- A transposase-driven reaction that fragments DNA into ~50–500 bp pieces and ligates adapters in a single step, compressing two library-prep stages into one.
- Hybrid capture
- Target enrichment using biotinylated probes that bind chosen regions (the panel genes), so sequencing reads are spent only where they are needed rather than across the whole genome.
- Coverage (depth)
- The number of independent reads spanning a given base. A 30× target means most bases are read at least thirty times, giving confidence in each genotype call.
- Mean genomic coverage
- The average read depth across the targeted footprint; validated hereditary-cancer assays have reported >99% sensitivity for germline SNVs, indels, and CNVs at ~35× mean coverage.
- [1]Scientific Reports 2016. A comprehensive custom panel design for routine hereditary cancer testing — Scientific Reports (srep39348). Notes ≥30× minimum coverage across regions of interest.↗
- [2]PMC 2020. Development and validation of a next-generation-sequencing-based 35-gene hereditary cancer panel. >99% of bases ≥30× vs ~94% for WES.↗
- [3]Illumina specifications. NovaSeq X Series specifications — 26–35 billion single reads per flow cell; 1.5B / 10B / 25B flow-cell options.↗
- [4]Illumina specifications. NovaSeq 6000 specifications — SP/S1/S2/S4 flow cells; up to ~20B paired-end reads per run.↗
The bench is only half the story
Sample collection and shipping decide what reaches the lab in the first place. See how a swab is preserved and transported before extraction even begins.