Finding an optimal sequencing strategy to detect short and long genetic variants in a human genome
Bibliographic record
Abstract
Abstract Advances in DNA sequencing have transformed genomics, enabling comprehensive insights into human genetic variation. While short-read sequencing (SRS) remains dominant due to its high accuracy and affordability, its limitations in complex genomic regions have spurred the adoption of long-read sequencing (LRS) platforms, such as those from Pacific Biosciences (PacBio) and Oxford Nanopore Technologies (ONT). Despite these advances, there is still a lack of systematic, large-scale benchmarking of variant calling performance across diverse platforms, variant types, sequencing depths, and genomic contexts. Here, we present a comprehensive benchmark of sequencing technologies and variant calling algorithms, evaluating their performance in detecting single nucleotide polymorphisms (SNPs), insertions/deletions (indels), and structural variants (SVs). We show that while SRS combined with DeepVariant or DRAGEN offers excellent small variant detection in well-mapped regions, LRS technologies significantly outperform SRS in complex regions and SV detection. PacBio achieves high SNP and smaller SV accuracy even at moderate coverage, while ONT excels in detecting large SVs. The SV callers Dysgu and SVIM emerged as top performers across LRS datasets. Our results highlight that no single platform is optimal for all variant types or regions: SRS remains optimal for high-throughput small variant detection in accessible regions, whereas LRS is critical for capturing SVs and resolving difficult-to-map loci. These findings offer practical guidance for selecting sequencing technologies, coverage and variant calling strategies tailored to specific research or clinical goals, contributing to more accurate and cost-effective genomic analyses. Highlights ● State-of-the-art short variant calling algorithms are highly comparable but focus on precision and sensitivity differently ● Short-read technologies outperform long-read technologies for small SNP and InDels, with the exception of difficult-to-map variants. ● To capture the majority of variants, a minimum coverage of 15x for PacBio, 20x for SRS, or 30x for ONT is required. However, optimal coverage depends on zygosity, variant type, and the region of interest. ● Long-read technologies outperform short-read technologies for all validation sets tested for deletions and insertions in all size categories.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".