Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 85% F1 (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Indel F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 87.3% Precision (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Indel Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 82.9% Recall (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Indel Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 96.4% F1 (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Overall F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 97.4% Precision (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Overall Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 95.4% Recall (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'Overall Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 98.1% F1 (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'SNP F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 98.9% Precision (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'SNP Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 97.3% Recall (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 10x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 10x, caller 'Clair3 pileup', column 'SNP Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 94.5% F1 (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Indel F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 95% Precision (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Indel Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 94.2% Recall (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Indel Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99% F1 (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Overall F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.1% Precision (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Overall Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99% Recall (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'Overall Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.7% F1 (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'SNP F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.8% Precision (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'SNP Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.7% Recall (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 20x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 20x, caller 'Clair3 pileup', column 'SNP Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 97.1% F1 (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Indel F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 97.2% Precision (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Indel Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 97% Recall (Indel; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Indel Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.5% F1 (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Overall F1-score' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.5% Precision (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Overall Precision' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.5% Recall (SNPs and indels; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'Overall Recall' Configuration: Clair3 pileup-only calls (Zheng et al. 2026 Supplementary Table 3) Protocol: HG003 PacBio Revio WGS small-variant accuracy against GIAB v4.2.1, Clair3 versus DeepVariant (Zheng et al. 2026 Supplementary Table 3) Dataset: GIAB HG003 whole genome, PacBio Revio (2022Q4 release), subsampled to 10x-50x 99.9% F1 (SNP; GIAB v4.2.1 HG003 confident regions; PacBio 30x)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 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 pileup, HG003 PacBio small-variant accuracy (Zheng et al. 2026)
model-execution-20261011-protocol-zheng2026-hg003-accuracy-pacbio
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 3 (image, page 3), platform 'PacBio Revio', read coverage 30x, caller 'Clair3 pileup', column 'SNP F1-score'