Configuration

DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)

Dragen v4.2 HS + Filters as run in the cited comparison.

Overview

Dragen v4.2 HS + Filters as run in the cited comparison.

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

Evaluations and results

2 evaluations · 28 results. Different protocols are not a single leaderboard.

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: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	100% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '1' / '0.5–1', column 'Precision HS-F (%)' (exons spanned block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	81% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '2–5' / '1–10', column 'Precision HS-F (%)' (exons spanned block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	68% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '>5' / '>10', column 'Precision HS-F (%)' (exons spanned block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	100% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '1' / '0.5–1', column 'Precision HS-F (%)' (CNV length block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	89% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '2–5' / '1–10', column 'Precision HS-F (%)' (CNV length block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: Coriell virtual gene-panel CNV precision by exon count and length (De La Vega et al. Table 1) Dataset: 25 Coriell cell lines with documented CNVs, 184-gene virtual panel	74% precision percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source HS-F on Coriell virtual panel cnv-20261009-protocol-delavega2025-coriell-virtual-panel Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Table 1, row '>5' / '>10', column 'Precision HS-F (%)' (CNV length block)
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.949 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, C6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 1-5 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	1 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, E6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 1-5 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	1 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, G6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 5-10 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	NA precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, I6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 5-10 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.68 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, K6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 10-50 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.182 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, M6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 10-50 kb'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.183 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, O6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 50 kb and over'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, Q6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 50 kb and over'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.776 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, S6; row 'Dragen v4.2 HS + Filters'; group 'deletions, all sizes'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.136 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, U6; row 'Dragen v4.2 HS + Filters'; group 'duplications, all sizes'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.621 precision fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, W6; row 'Dragen v4.2 HS + Filters'; group 'deletions and duplications combined'; column Precision
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.822 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, B6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 1-5 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.25 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, D6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 1-5 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.926 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, F6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 5-10 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	NA recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, H6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 5-10 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	0.895 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, J6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 10-50 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	1 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, L6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 10-50 kb'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	1 recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, N6; row 'Dragen v4.2 HS + Filters'; group 'deletions, 50 kb and over'; column Sensitivity
Configuration: DRAGEN 4.2 high-sensitivity mode with custom artifact filters (HS-F)	Protocol: HG002 exon-overlap CNV benchmark by event type and length (De La Vega et al. Table S3) Dataset: HG002 50x PCR-free WGS with GIAB SV v0.6 CNV truth and synthetic gene models	NA recall fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Results 3.1 paragraph 5 says CNVnator and Cue were unable to detect events of 1-5 kb, but Supplemental Table 3 prints CNVnator 1-5 kb deletion sensitivity 0.3098 and precision 0.4207 (B9, C9). Results 3.1 paragraph 3 says Delly had the lowest precision, but Table S3 prints lower combined precision for Lumpy (W13 0.0107) than Delly (W11 0.1902). The prose may describe Figures 1 and 3 rather than the table; this was not checked. Table values are recorded.Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Results 3.1 paragraphs 3 and 5 versus Supplemental_Table_3.xlsx B9, C9, W11, W13 Methods, coverage and source Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark cnv-20261009-protocol-delavega2025-hg002-exon-overlap Aggregation: Not reported Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications; De La Vega et al. 2025, Supplemental Table 3 · Supplemental Table 3, P6; row 'Dragen v4.2 HS + Filters'; group 'duplications, 50 kb and over'; column Sensitivity

Source checking is not independent reproduction. Release 2026-10-09-9307685239b3.

Use this model

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 2026-10-09-9307685239b3
Property and statement	Original source and location	Review and provenance

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

All evidence (JSONL) (gzip)How to read the evidence

Sources and history

Release 2026-10-09-9307685239b3 · Record review: source checked

2 source records and release history

Benchmarking of germline copy number variant callers from whole genome sequencing data for clinical applications · Original source · Bioinformatics Advances 5(1):vbaf071, published 2025-04-10; PMC12005901 full-text XML
De La Vega et al. 2025, Supplemental Table 3 · Original source · Supplemental_Table_3.xlsx inside vbaf071_supplementary_data.zip

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: cnv-20261009-config-delavega2025-dragen42-hs-filters

areas: dna-genomes
contexts: clinical_research
method types: conventional_pipeline
reported name: Dragen v4.2 HS + Filters
version: 4.2
protocol: High-sensitivity mode (-sv-cnv-enable-high-sensitivity-mode=true) followed by the custom RTG vcffilter scheme in Supplementary File S1
foundation model eligible: false

Related records

configuration of: DRAGEN
subject: post_processing: cnv-20261009-config-delavega2025-dragen42-hs-filters
system: Dragen v4.2 HS + Filters on HG002 exon-overlap benchmark
system: HS-F on Coriell virtual panel

Suggest a correction