Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 396 Runtime (Clair3 GPU workflow, full-alignment calling stage; HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'Full-alignment calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 72 Runtime (Clair3 GPU workflow, LongPhase variant phasing stage; HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'LongPhase for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 500 Runtime (Clair3 GPU workflow, pileup calling stage; HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'Pileup calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 1060 Runtime (Clair3 GPU workflow, total; HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'Total runtime' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 723 Runtime (Clair3 GPU workflow, WhatsHap variant phasing stage (alternative to LongPhase); HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'or WhatHap for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 87 Runtime (Clair3 GPU workflow, workflow control; HG003 ONT 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 30x, column 'Workflow control' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 538 Runtime (Clair3 GPU workflow, full-alignment calling stage; HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'Full-alignment calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 93 Runtime (Clair3 GPU workflow, LongPhase variant phasing stage; HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'LongPhase for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 606 Runtime (Clair3 GPU workflow, pileup calling stage; HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'Pileup calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 1260 Runtime (Clair3 GPU workflow, total; HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'Total runtime' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 801 Runtime (Clair3 GPU workflow, WhatsHap variant phasing stage (alternative to LongPhase); HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'or WhatHap for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, ONT R10.4.1 5 kHz (EPI2ME giab-2025.01), subsampled to 10x-50x 22 Runtime (Clair3 GPU workflow, workflow control; HG003 ONT 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 ONT (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'ONT r10.4.1 5kHz', read coverage 50x, column 'Workflow control' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 235 Runtime (Clair3 GPU workflow, full-alignment calling stage; HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'Full-alignment calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 42 Runtime (Clair3 GPU workflow, LongPhase variant phasing stage; HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'LongPhase for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 359 Runtime (Clair3 GPU workflow, pileup calling stage; HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'Pileup calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 706 Runtime (Clair3 GPU workflow, total; HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'Total runtime' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 653 Runtime (Clair3 GPU workflow, WhatsHap variant phasing stage (alternative to LongPhase); HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'or WhatHap for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 70 Runtime (Clair3 GPU workflow, workflow control; HG003 PacBio 30x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 30x, column 'Workflow control' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 282 Runtime (Clair3 GPU workflow, full-alignment calling stage; HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'Full-alignment calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 67 Runtime (Clair3 GPU workflow, LongPhase variant phasing stage; HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'LongPhase for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 357 Runtime (Clair3 GPU workflow, pileup calling stage; HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'Pileup calling' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 739 Runtime (Clair3 GPU workflow, total; HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'Total runtime' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 826 Runtime (Clair3 GPU workflow, WhatsHap variant phasing stage (alternative to LongPhase); HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'or WhatHap for variant phasing' Configuration: Clair3 v1.2.0 GPU mode (Zheng et al. 2026) Protocol: Clair3 GPU workflow stage runtimes on HG003 ONT and PacBio WGS (Zheng et al. 2026 Supplementary Table 1) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 33 Runtime (Clair3 GPU workflow, workflow control; HG003 PacBio 50x)second · lower
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons Supplementary Table 2 runtimes differ from the article's Results for what appear to be the same HG003 comparisons. ONT at 30x and 50x: Clair3 GPU 17 and 20 minutes in the table, 16 and 15 in Results 3.1.1; DeepVariant GPU 48 and 82 in the table, 106 to 128 in the text. PacBio Revio at 30x and 50x: Clair3 GPU 12 and 12 in the table, 23 and 21 in Results 3.1.2; DeepVariant GPU 40 and 53 in the table, 93 and 118 in the text. The table's Clair3 PacBio runtimes equal the macOS M3 Ultra runtime in Results 3.1.3, not the Linux RTX 4090 runtimes. The table and the Supplementary Notes commands give DeepVariant v1.10-beta; Results 3.1 say v1.8. Supplementary Table 3 DeepVariant ONT SNP precision, recall and F1 at 10x and 20x (97.58%, 92.35%, 94.89% and 99.32%, 99.02%, 99.17%) differ from Results 3.2 (97.34%, 92.17%, 94.68% and 98.75%, 98.79%, 98.79%), while the Clair3 values agree. Results 3.2 cite Supplementary Table 2 for the HG001 to HG007 comparison, which is Supplementary Table 4. The hardware, DeepVariant version and runs behind Supplementary Tables 2 and 3 are therefore uncertain. Supplementary Table 1 agrees with Results 3.1.4 except WhatsHap ONT 30x (723 seconds in the table, 724 in the text). Table values are recorded; text values are not.Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Tables 1 to 3 (pages 2 and 3) versus article Results 3.1.1 to 3.1.4 and 3.2
Methods, coverage and source Clair3 v1.2.0 GPU stage runtimes, HG003 PacBio (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-clair3-stage-runtime
Aggregation: Not reported
Accelerated long-read variant calling with Clair3 for whole-genome sequencing ; Accelerated long-read variant calling with Clair3: Supplementary Data · Supplementary Table 1, sample HG003, platform 'PacBio Revio', read coverage 50x, column 'Workflow control'