Configuration

ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)

ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi: Model trained on synthetic tumours only (GIAB HG001/HG002 for ONT; HG003/HG004 for PacBio and Illumina); default panels of normals (gnomAD r2.1, dbSNP v138, 1000 Genomes PoN, CoLoRSdb) and hard filters; short-read runs omit long-read-specific filters.

2 evaluations · 14 results

Overview

An explanatory profile has not yet been reviewed for this record.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Evaluations and results

2 evaluations · 14 results. Different protocols are not a single leaderboard.

Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)

Sorted by AUPRC (indel only, tumour-only, 50x tumour coverage) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Protocol and dataset	AUPRC (indel only, tumour-only, 50x tumour coverage)	AUPRC (SNV only, tumour-only, 50x tumour coverage)	F1 (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	F1 (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	False negatives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	False negatives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	False positives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	False positives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	Precision (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	Precision (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	Recall (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	Recall (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	True positives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold)	True positives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold)
ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)Evaluation	PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a)	0.189	Not reported	0.262	Not reported	308	Not reported	1,396	Not reported	0.178	Not reported	0.495	Not reported	302	Not reported
ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)Evaluation	PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a)	Not reported	0.644	Not reported	0.785	Not reported	2,901	Not reported	16,917	Not reported	0.681	Not reported	0.926	Not reported	36,185

All 14 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.189 AUPRC (indel only, tumour-only, 50x tumour coverage) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'AUPRC'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.262 F1 (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'F1-score'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	308 False negatives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'FN'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	1,396 False positives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'FP'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.178 Precision (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'Precision'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.495 Recall (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'Recall'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic indels (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	302 True positives (indel only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-indel Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'Indel', column 'TP'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.644 AUPRC (SNV only, tumour-only, 50x tumour coverage) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'AUPRC'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.785 F1 (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'F1-score'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	2,901 False negatives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'FN'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	16,917 False positives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'FP'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.681 Precision (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'Precision'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	0.926 Recall (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'Recall'
Configuration: ClairS-TO SS v0.4.0, tumour-only, PacBio HiFi (Chen et al. 2025)	Protocol: PacBio HiFi COLO829 tumour-only, 50x, somatic SNVs (Chen et al. 2025 Supplementary Table 7a) Dataset: COLO829 tumour PacBio HiFi Revio WGS (ClairS-TO benchmark)	36,185 True positives (SNV only, tumour-only, 50x tumour coverage, at the best-F1 threshold) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025) somatic-clairs-to-20261011-protocol-pacbio-colo829-snv Aggregation: Not reported ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling; ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines); ClairS-TO Source Data workbook · Supplementary Table 7a, Dataset 'COLO829', Caller 'ClairS-TO SS', Tumor coverage '50x', Type 'SNV', column 'TP'

Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.

Use this configuration

How it works, versions and access

Strengths, limitations and unresolved questions

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

3 source records and release history

ClairS-TO: a deep-learning method for long-read tumor-only somatic small variant calling · Original source · Nature Communications, published 2025-10-31; PMC12579226 full-text XML
ClairS-TO Supplementary Information (Supplementary Tables 1-8 and command lines) · Original source · Supplementary Information PDF (41467_2025_64547_MOESM1_ESM.pdf) of the published article
ClairS-TO Source Data workbook · Original source · Source Data file (41467_2025_64547_MOESM6_ESM.xlsx) of the published article

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: somatic-clairs-to-20261011-config-clairs-to-ss-pacbio-hifi

areas: dna-genomes
contexts: clinical_research
method types: supervised_machine_learning
reported name: ClairS-TO SS
version: v0.4.0
foundation model eligible: false
parameters: Model trained on synthetic tumours only (GIAB HG001/HG002 for ONT; HG003/HG004 for PacBio and Illumina); default panels of normals (gnomAD r2.1, dbSNP v138, 1000 Genomes PoN, CoLoRSdb) and hard filters; short-read runs omit long-read-specific filters
access reuse: Public open-source software and released models
source locator: Supplementary Information 'Command lines used'; Supplementary Tables 2 and 7 row labels
provenance note: ClairS-TO SS as run on PacBio HiFi tumour-only data in the cited comparison.

Related records

configuration of: ClairS-TO
system: ClairS-TO SS on pacbio-colo829 tumour-only Indel (Chen et al. 2025)
system: ClairS-TO SS on pacbio-colo829 tumour-only SNV (Chen et al. 2025)

Suggest a correction