Configuration

DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)

DeepSomatic 1.10.0, bundled WGS model: google/deepsomatic:1.10.0, digest sha256:6cd98c41781cdc1a2916d8029bedef70a57b87d41f933bd4cf259c80f291e27b6; bundled whole-genome model (manifest SHA-256 3d909228d0224370959222cfc7f67d92ab76436d4455be92b9fb90cdb9ff7a66); no retraining or tuning; PASS calls.

4 evaluations · 24 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

4 evaluations · 24 results. Different protocols are not a single leaderboard.

Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3

Sorted by F1 (indel only, HG008 primary region) (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	F1 (indel only, HG008 primary region)	F1 (SNV only, HG008 primary region)	F1 (SNVs and indels, HG008 all-benchmark region)	F1 (SNVs and indels, HG008 primary region)	False negatives (indel only, HG008 primary region)	False negatives (SNV only, HG008 primary region)	False negatives (SNVs and indels, HG008 all-benchmark region)	False negatives (SNVs and indels, HG008 primary region)	False positives (indel only, HG008 primary region)	False positives (SNV only, HG008 primary region)	False positives (SNVs and indels, HG008 all-benchmark region)	False positives (SNVs and indels, HG008 primary region)	Precision (indel only, HG008 primary region)	Precision (SNV only, HG008 primary region)	Precision (SNVs and indels, HG008 all-benchmark region)	Precision (SNVs and indels, HG008 primary region)	Recall (indel only, HG008 primary region)	Recall (SNV only, HG008 primary region)	Recall (SNVs and indels, HG008 all-benchmark region)	Recall (SNVs and indels, HG008 primary region)	True positives (indel only, HG008 primary region)	True positives (SNV only, HG008 primary region)	True positives (SNVs and indels, HG008 all-benchmark region)	True positives (SNVs and indels, HG008 primary region)
DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)Evaluation	HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL')	0.69	Not reported	Not reported	Not reported	2324	Not reported	Not reported	Not reported	28	Not reported	Not reported	Not reported	0.989	Not reported	Not reported	Not reported	0.529	Not reported	Not reported	Not reported	2612	Not reported	Not reported	Not reported
DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)Evaluation	HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4)	Not reported	Not reported	0.776	Not reported	Not reported	Not reported	6589	Not reported	Not reported	Not reported	133	Not reported	Not reported	Not reported	0.989	Not reported	Not reported	Not reported	0.638	Not reported	Not reported	Not reported	11618	Not reported
DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)Evaluation	HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1)	Not reported	Not reported	Not reported	0.897	Not reported	Not reported	Not reported	2510	Not reported	Not reported	Not reported	124	Not reported	Not reported	Not reported	0.989	Not reported	Not reported	Not reported	0.821	Not reported	Not reported	Not reported	11504
DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)Evaluation	HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV')	Not reported	0.984	Not reported	Not reported	Not reported	186	Not reported	Not reported	Not reported	96	Not reported	Not reported	Not reported	0.989	Not reported	Not reported	Not reported	0.98	Not reported	Not reported	Not reported	8892	Not reported	Not reported

All 24 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: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.776 F1 (SNVs and indels, HG008 all-benchmark region) fraction · higher Uncertainty: 95% CI 0.7689 to 0.7822. Genomic block bootstrap over 5-Mb blocks, linear percentile interval; prespecified, but executed after the primary external evaluation; spatial variation within the benchmark region, not between tumours Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'F1'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	6590 False negatives (SNVs and indels, HG008 all-benchmark region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'FN'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	133 False positives (SNVs and indels, HG008 all-benchmark region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'FP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.989 Precision (SNVs and indels, HG008 all-benchmark region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'Precision'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.638 Recall (SNVs and indels, HG008 all-benchmark region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'Recall'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 4) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	11600 True positives (SNVs and indels, HG008 all-benchmark region) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 all-benchmark region (sensitivity analysis), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-all-region-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table4_HG008_sensitivity_all_region.tsv, row 'DeepSomatic', column 'TP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.69 F1 (indel only, HG008 primary region) fraction · higher Uncertainty: 95% CI 0.676907 to 0.701908. Genomic block bootstrap over 5-Mb blocks, linear percentile interval; spatial variation within the benchmark region, not between tumours Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'F1'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	2320 False negatives (indel only, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'FN'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	28 False positives (indel only, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'FP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.989 Precision (indel only, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'Precision'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.529 Recall (indel only, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'Recall'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), indels, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'INDEL') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	2610 True positives (indel only, HG008 primary region) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), indels (Hasan 2026) somatic-hg008-20261011-protocol-primary-indel Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'INDEL', column 'TP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.897 F1 (SNVs and indels, HG008 primary region) fraction · higher Uncertainty: 95% CI 0.892866 to 0.901452. Genomic block bootstrap over 5-Mb blocks, linear percentile interval; spatial variation within the benchmark region, not between tumours Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'F1'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	2510 False negatives (SNVs and indels, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'FN'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	124 False positives (SNVs and indels, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'FP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.989 Precision (SNVs and indels, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'Precision'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.821 Recall (SNVs and indels, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'Recall'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs and indels combined, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 1) Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	11500 True positives (SNVs and indels, HG008 primary region) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs and indels combined (Hasan 2026) somatic-hg008-20261011-protocol-primary-overall Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table1_HG008_primary_metrics.tsv, row 'DeepSomatic', column 'TP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.984 F1 (SNV only, HG008 primary region) fraction · higher Uncertainty: 95% CI 0.982090 to 0.986558. Genomic block bootstrap over 5-Mb blocks, linear percentile interval; spatial variation within the benchmark region, not between tumours Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'SNV', column 'F1'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	186 False negatives (SNV only, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'SNV', column 'FN'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	96 False positives (SNV only, HG008 primary region) count · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'SNV', column 'FP'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.989 Precision (SNV only, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'SNV', column 'Precision'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	0.98 Recall (SNV only, HG008 primary region) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class 'SNV', column 'Recall'
Configuration: DeepSomatic 1.10.0, bundled WGS model (Hasan 2026)	Protocol: HG008 primary region (nogermlineinterference), SNVs, Mutect2, Strelka2, DeepSomatic and ensembles (Hasan 2026 Table 2, Class 'SNV') Dataset: NIST/GIAB HG008 tumour-normal WGS (HG008-T pancreatic cancer cell line, matched normal), HG008-T draft somatic small-variant benchmark v0.3	8890 True positives (SNV only, HG008 primary region) count · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DeepSomatic on HG008 primary region (nogermlineinterference), SNVs (Hasan 2026) somatic-hg008-20261011-protocol-primary-snv Aggregation: Not reported Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling; AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Zenodo tables/Table2_HG008_variant_class_metrics.tsv, row 'DeepSomatic', Class '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

2 source records and release history

Prospectively frozen external benchmarking reveals benchmark-context dependence in somatic variant calling · Original source · Research Square preprint rs-11210923 version 1, posted 2026-10-05; manuscript PDF (16 pages)
AmbioSpire P003/T001 publication source data and reproducibility package v1.0.0 (Zenodo) · Original source · Zenodo record 22998005, version 1.0.0, published 2026-09-27; file P003-T001-publication-data-v1.0.0-zenodo-final.zip (1,934,562 bytes, md5 a1fdab6108197602f07f72db77d9f996)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: somatic-hg008-20261011-config-deepsomatic

areas: dna-genomes
contexts: clinical_research
method types: supervised_machine_learning
reported name: DeepSomatic
foundation model eligible: false
parameters: google/deepsomatic:1.10.0, digest sha256:6cd98c41781cdc1a2916d8029bedef70a57b87d41f933bd4cf259c80f291e27b6; bundled whole-genome model (manifest SHA-256 3d909228d0224370959222cfc7f67d92ab76436d4455be92b9fb90cdb9ff7a66); no retraining or tuning; PASS calls
access reuse: Public open-source software
source locator: Online Methods 'Variant callers and retained software identities' and 'Consensus definitions and metrics'; Zenodo metadata/software_versions.tsv
version: 1.10.0
provenance note: DeepSomatic as run in the cited HG008 evaluation.

Related records

Suggest a correction