Back
PubMedLong-read sequencingClinical pipelineBenchmark

Sensitivity of HiFi long-read genome sequencing for difficult-to-detect pathogenic variants when applied to real-world clinical laboratory samples.

Devaney JM, Chong JX, Lopes PC, Noya J, Berlyoung AS, Yusuff S, et al.Am J Hum Genet 2026 · May 2026
Relevance score
9/10
Disease / domain
Difficult-to-detect pathogenic variants (repeat expansions, deep intronic, structural variants, mosaicism)
Source
PubMed
PMID 42049032
Share on LinkedIn

Variant / mechanism

HiFi long-read genome sequencing (PacBio ~30×) in real-world clinical laboratory conditions

Summary

HiFi long-read genome sequencing (PacBio, ~30×) was applied to 191 probands enriched for difficult-to-detect pathogenic variants across a real clinical laboratory. The automated pipeline detected 479 of 481 expected variants (99.6%), including repeat expansions, deep intronic variants, complex SVs, and mosaic variants — even from suboptimal samples such as buccal swabs and low-molecular-weight DNA. Only two mosaic variants at low allele fractions (trisomy 18 at 23% and a deletion at 13%) were missed.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Analysis

This study provides one of the strongest clinical-grade validations of HiFi long-read WGS in a diagnostic laboratory setting. The 99.6% detection rate across all difficult-to-detect variant classes — on imperfect real-world samples — is a critical milestone for moving beyond the sequential multi-assay paradigm. The residual blind spot for low-level mosaicism (<23%) is shared with current technologies. Single first-tier long-read WGS now appears within practical reach.

Why this score?

Clinical impact : 3/3 · Evidence strength : 3/3 · Novelty : 1/2 · Sample size : 1/1 · Journal quality : 1/1 → Total : 9/10

Keywords

long-read WGSHiFi PacBiodifficult-to-detect variantsclinical sequencingmosaicism
Weekly report in your inbox

Every Wednesday · Annotated selection · Free · Unsubscribe anytime