ClairS: a deep-learning method for long-read tumor–normal pair somatic small variant calling
- 1School of Computing and Data Science, The University of Hong Kong, Hong Kong, China
- 2School of Computing and Data Science, The University of Hong Kong, Hong Kong, China. [email protected]
Abstract
Somatic variant discovery in tumors is crucial for clinical analysis, yet most existing methods are designed for short-read sequencing, with few developed specifically for long reads. Here we present Clair-Somatic (ClairS), a deep-learning-based somatic small-variant caller designed for long-read tumor-normal pairs. Trained on synthetic somatic variants with diverse coverages and variant allele fractions, ClairS accurately detects a wide range of somatic variants. Using the Nanopore Q20+ HCC1395-HCC1395BL dataset at 50/25× tumor/normal coverage, ClairS achieved F1 scores of 89.83% for single-nucleotide variations and 73.38% for indels; augmenting training with real cancer cell lines improved performance to 96.19% and 79.67%, respectively. Our findings indicate that improved read phasing enabled by long-read sequencing is key to accurate single-nucleotide variation detection, especially at low variant allele fractions. Through experiments across varied coverage, purity, contamination levels, multiple platforms and real cancer cell lines, we demonstrate that ClairS is a robust and reliable caller. ClairS is open source and available at https://github.com/HKU-BAL/ClairS .
Related
- Accurate calling of low-frequency somatic mutations by sample-specific modeling of error rates
- Accurate prediction of CRISPR editing outcomes in somatic cell lines and zygotes with few-shot learning with inDecay.
- iGC: Antibody Optimization via Reinforcement Learning over a Learned Somatic Hypermutation Prior
- From conservation scores to language models: a feature engineering perspective on machine learning for genetic variant interpretation
- Controlling for DNA dosage with whole-genome sequencing improves ATAC-seq peak calling
BibTeX
@article{zheng2026clairs,
title = {{ClairS: a deep-learning method for long-read tumor–normal pair somatic small variant calling}},
author = {Zhenxian Zheng and Lei Chen and Junhao Su and Xian Yu and Minggao He and Yan-Lam Lee and Tak‐Wah Lam and Ruibang Luo},
journal = {Nature Methods},
year = {2026},
month = jul,
doi = {10.1038/s41592-026-03152-4},
url = {https://doi.org/10.1038/s41592-026-03152-4},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - ClairS: a deep-learning method for long-read tumor–normal pair somatic small variant calling
AU - Zhenxian Zheng
AU - Lei Chen
AU - Junhao Su
AU - Xian Yu
AU - Minggao He
AU - Yan-Lam Lee
AU - Tak‐Wah Lam
AU - Ruibang Luo
PY - 2026
DA - 2026/07/01
JO - Nature Methods
DO - 10.1038/s41592-026-03152-4
UR - https://doi.org/10.1038/s41592-026-03152-4
AB - Somatic variant discovery in tumors is crucial for clinical analysis, yet most existing methods are designed for short-read sequencing, with few developed specifically for long reads. Here we present Clair-Somatic (ClairS), a deep-learning-based somatic small-variant caller designed for long-read tumor-normal pairs. Trained on synthetic somatic variants with diverse coverages and variant allele fractions, ClairS accurately detects a wide range of somatic variants. Using the Nanopore Q20+ HCC1395-HCC1395BL dataset at 50/25× tumor/normal coverage, ClairS achieved F1 scores of 89.83% for single-nucleotide variations and 73.38% for indels; augmenting training with real cancer cell lines improved performance to 96.19% and 79.67%, respectively. Our findings indicate that improved read phasing enabled by long-read sequencing is key to accurate single-nucleotide variation detection, especially at low variant allele fractions. Through experiments across varied coverage, purity, contamination levels, multiple platforms and real cancer cell lines, we demonstrate that ClairS is a robust and reliable caller. ClairS is open source and available at https://github.com/HKU-BAL/ClairS .
SN - 1548-7091
SN - 1548-7105
ER -