Show Notes
Jang Y et al., The American Journal of Human Genetics - Every cell in the body picks up its own small set of DNA changes over a lifetime, and many of these somatic mutations are carried by only a tiny fraction of cells. This study, part of the SMaHT Network, asks how reliably standard bulk whole-genome sequencing can find mutations present at a variant allele fraction below 1%. The team built a ground truth from one donor's skin: clonal stem cell lines grown from single fibroblasts defined real mutations, and lines from other skin biopsies of the same donor defined mutations that should be absent. With about 1,360× Illumina coverage confirmed by PacBio and AVITI, point mutations below 1% were detected accurately in the easily mapped parts of the genome. Key terms: somatic mosaicism, whole-genome sequencing, variant allele fraction, benchmarking, SMaHT.
Study Highlights:
Primary fibroblasts from a 29-year-old donor contained five subclones, captured as clonal iPSC lines, giving benchmark mutations from 0.1% to 25% variant allele fraction, while iPSC lines from other biopsies of the same donor supplied negative controls; both sets held about 11,000 SNVs and 2,800 indels. Using the negative controls, PacBio and AVITI had the lowest substitution error rates in easy regions, 0.003% and 0.005%, and mismatch rates rose by 1 to 2 orders of magnitude in harder regions. In fibroblasts, more than 99.5% of bulk-only calls above 1% were confirmed by PacBio or AVITI, the estimated false-positive rate was near zero above 0.8% and about 11% below 0.5%, and homopolymer indels could be detected reliably only above 3%. In postmortem lung and colon, combining about 200× from two tissues found 13 additional shared mutations, a 145% gain in sensitivity below 3%, all confirmed by PacBio or single-cell data.
Conclusion:
Ultra-deep bulk whole-genome sequencing can detect point mutations below 1% allele fraction with high accuracy, provided calling is restricted to easily mapped regions and sample quality is checked, and pooling coverage across tissues from one donor recovers shared developmental mutations. The benchmark comes from a single donor's fibroblasts, confirmation below 0.5% was limited by long-read and AVITI coverage, most indels sit in homopolymers or hard regions where detection remains unreliable, and postmortem samples needed extra filtering.
Article title:
Accurate detection of sub-1% frequency somatic mutations by whole-genome sequencing
First author:
Jang Y
Journal:
The American Journal of Human Genetics
DOI:
10.1016/j.ajhg.2026.09.010
Reference:
Jang Y., Gardiner S., Wang Y., Natu A., Dehankar M.K., Tomasini L., et al. Accurate detection of sub-1% frequency somatic mutations by whole-genome sequencing. The American Journal of Human Genetics. (2026). https://doi.org/10.1016/j.ajhg.2026.09.010
License:
This episode is based on an open-access article published under the Creative Commons Attribution 4.0 International License (CC BY 4.0) – https://creativecommons.org/licenses/by/4.0/
Support:
Base by Base is independent and ad-free — no sponsors, no paywall. If an episode was worth your time, chip in and keep the papers audited and the original songs coming:
❤️ Support monthly: https://buy.stripe.com/cNifZhclVebvagk2JDgEg01
☕ One-time donation: https://donate.stripe.com/7sY4gz71B2sN3RWac5gEg00
▶️ Watch with chapters on YouTube: https://www.youtube.com/@basebybase
More at basebybase.com
On PaperCast Base by Base you'll discover the latest in genomics, functional genomics, structural genomics, and proteomics.
Episode link: https://basebybase.com/episodes/detecting-sub-1-percent-somatic-mutations-wgs
QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2026-10-18.
QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music
- transcript coverage: Context on somatic mutations, VAF and SMaHT; prior benchmark weaknesses; study design with iPSC lines; platform error rates; SNV and indel detection; phasing-based false-positive rates; caller complementarity; coverage effects; postmortem filters; joint tissue analysis; implications and limits.
- transcript topics: Somatic mutations and variant allele fraction; SMaHT Network goals; Benchmark built from one donor's fibroblasts and iPSC lines; Sequencing platform error rates; Sub-1% SNV detection and confirmation; Read-backed phasing false-positive estimates
QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0
Metadata Audited:
- article_doi
- article_title
- article_journal
- license
Factual Items Audited:
- SMaHT: 10-15 tissues from 150 non-diseased donors at about 300x
- Donor LB, 29-year-old male, 25 clonal iPSC lines from six biopsies
- About 2,000 SNVs and 550 indels per line; 99.5% and 98.5% PacBio validation
- Benchmark and negative sets of about 11,000 SNVs and 2,800 indels; VAF 0.1%-25%
- About 1,360x Illumina, about 380x PacBio, about 220x AVITI
- Easy regions about 74% of GRCh38
QC result: Pass.
