Evidence summary: diagnostic genomics model execution: value
Claimed value of Evidence summary: diagnostic genomics model execution
Eight published execution benchmarks are linked, alongside the DRAGEN 4.2 developer total runtime for HG002 (1,838.5 seconds); two are withheld pending source concerns (see the end of this summary). Of the six active benchmarks, five measure conventional variant-calling pipelines and one measures inference of a genomic foundation model on microbial genomes, outside any diagnostic workflow. Only the foundation-model benchmark prints spread over repeated runs (five trials); none reruns a variant-calling pipeline to measure concordance of the calls, none prints variant-calling accuracy in the same table as runtime, and hardware differs within most comparisons, so differences mix software and hardware. On ten WGS samples run by a University of Oslo and Oslo University Hospital group with no declared competing interests (Samarakoon et al. 2025), read mapping plus HaplotypeCaller for NA12878 (47.55x) took 2,317.7 minutes on a CPU-only GATK pipeline, 123.45 minutes with Parabricks on four L4 GPUs, 78.12 on four A100 GPUs, 54.22 on eight H100 GPUs and 60.29 with DRAGEN v4.2. On the same H100 machine, Parabricks DeepVariant took 14.58 minutes for NA12878 with the version the pipeline used (printed 4.1.0-1; Methods say 4.3.0-1) and 16.05 minutes with the updated version (printed 4.3.0-1; Discussion says 4.3.1-1). On one HG002 30x genome on Google Cloud (O'Connell et al. 2023; the authors' employer funded the study and is an NVIDIA, AWS and Google alliance partner), HaplotypeCaller took 38.8 hours and cost 67.9 USD on a 32-vCPU machine and 0.59 hours and 17.5 USD on eight A100 GPUs, while the somatic caller LoFreq cost 8.1 USD on the CPU machine and 19 to 30.1 USD on 2 to 8 A100 GPUs. On Google Cloud machines of similar hourly cost (Franzoso et al. 2025, no declared conflicts), Sentieon DNASeq took 0:14:00 to 0:16:00 per exome at 0.82 to 1.03 USD and 3:00:00 to 3:50:00 per genome at 8.02 to 10.7 USD, and Parabricks Germline on one T4 GPU took 0:10:00 to 0:14:00 per exome at 0.71 to 0.93 USD and 4:06:00 to 4:40:00 per genome at 8.13 to 10.6 USD; costs exclude the Sentieon licence, and two genome runtime cells that cannot be genome runtimes are disputed and left out of these ranges. On human WGS at five read depths on the HPC nodes of a regional genome project (Marangoni et al. 2026; the GeNePi developers, no declared competing interests; one unnamed sample per depth), a 30x genome took 17h with the CPU GATK equivalent and 1h 36m, 42 m and 30m with Parabricks v4.3.0 on two V100, two A100 and four A100 GPUs, and a 100x genome 60h, 6h 30m, 2h 30m and 2h 06m. On one 120-vCPU Azure instance (Hu et al. 2026; all authors employed by Sentieon), DNAscope took 18.9 minutes for 30x Illumina, DNAscope LongRead 41.2 minutes for 30x PacBio HiFi and DNAscope Hybrid 85.8 minutes for 10x PacBio HiFi plus 35x Illumina, at 1.1, 2.5 and 5.2 USD on demand excluding the licence; each pipeline had different input data, so the three are not a ranking. For GenomeOcean on one H100 (Yu et al. 2026; two authors are at the DOE Joint Genome Institute, which develops GenomeOcean; GTDB bacterial genomes; mean and standard deviation over five trials), embedding extraction with the 4B model ran at 21,386 ± 126 input tokens per second with 72.14 GB peak GPU memory in BF16 and 22,401 ± 204 with 67.91 GB with FP8 weights, and generation ran at 1818.79 ± 7.95 generated tokens per second in BF16 and 2169.04 ± 24.30 with FP8 weights, activations and KV cache. No active linked result measures variant-calling accuracy against a truth set; the GenomeOcean ARI values score metagenomic binning against GTDB species labels. The two withheld studies are Zheng et al. 2026 (Clair3 developers; Clair3 and DeepVariant on HG003 long reads), whose supplementary runtimes, DeepVariant version and DeepVariant accuracy differ from its Results text and whose accuracy uses one GIAB sample's v4.2.1 confident regions, which leave out the hardest regions; and Aspidova et al. 2026 (ModernGENA and GENA-LM developers; four DNA language model families on one A100), whose printed tokens per second are far too low for its batch sizes. Apart from the five GenomeOcean trials, each comparison prints one value per sample and set-up on at most ten samples, and costs are list prices at the time of each study.
Replace the use-case summary with the reviewer's text after the 2026-10-11 execution literature pass added four active and two withheld studies. Historical releases retain the previous value.
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
70 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Claimed field attributes.value Context-only references | DRAGEN supplementary tables 1–17 amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: MOESM3 XLSX retrieved 2026-10-07 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record Format: curator_factual_receipt |
| Claimed field attributes.value Context-only references | Rapid NGS Analysis on Google Cloud Platform: Performance Benchmark and User Tutorial amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Clinical and Translational Science 18(11):e70416, published 2025-11-18; PMC12627752 full-text XML | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field attributes.value Context-only references | Franzoso et al. 2025, Table S2 (runtime and costs per sample) and Table S1 (sample identifiers) amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: CTS-18-e70416-s002.xlsx (Table S2) inside the Europe PMC supplementary files bundle for PMC12627752 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of CTS-18-e70416-s002.xlsx. Table S1 (CTS-18-e70416-s001.xlsx) from the same bundle has SHA-256 b221919e435217c607ff426d80fc5718c7f4d3e0fbaa6261cf13e8c5b7ae0f4e and is used only to assert sample identifiers. The bundle is assembled per request, so only member hashes are pinned. Archive member: CTS-18-e70416-s002.xlsx |
| Claimed field attributes.value Context-only references | Accelerating genomic workflows using NVIDIA Parabricks amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: BMC Bioinformatics 24:221, published 2023-05-31; PMC10230726 full-text XML | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field attributes.value Context-only references | Benchmarking accelerated next-generation sequencing analysis pipelines amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Bioinformatics Advances 5(1):vbaf085, published 2025-05-15; PMC12092081 full-text XML | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field attributes.value Context-only references | Samarakoon et al. 2025, supplementary materials (Tables S1-S4, supplementary texts) amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 'Publication-ready_Supplementary materials-20250404.pdf' inside vbaf085_supplementary_data.zip | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of the PDF. The publisher zip vbaf085_supplementary_data.zip that contains it has SHA-256 a828608e459ec1fe57828e9b4fbce228e9507b8686eecedac323f45fd4c3cc07. Retrieved inside the Europe PMC supplementaryFiles zip, which is assembled per request; the publisher's inner zip and the PDF it holds are pinned. Archive member: Publication-ready_Supplementary materials-20250404.pdf |
| Claimed field attributes.value Context-only references | A novel and accelerated method for integrated alignment and variant calling from short and long reads amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Frontiers in Bioinformatics 5:1691056, published 2026-01-05; PMC12813096 full-text XML | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field attributes.value Context-only references | GeNePi: a graphics processing unit enhanced next-generation bioinformatics pipeline for whole-genome sequencing analysis amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Briefings in Bioinformatics 27(1), published 2026-01-25; PMC12832024 full-text XML | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Claimed field attributes.value Context-only references | GeNePi: Supplementary Information amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: genepi_supplinfo_rev_bbag001.pdf from the Europe PMC supplementary files of PMC12832024 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record Archive member: genepi_supplinfo_rev_bbag001.pdf |
| Claimed field attributes.value Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
10 source records and release history
- DRAGEN supplementary tables 1–17 · Original source · MOESM3 XLSX retrieved 2026-10-07
- Rapid NGS Analysis on Google Cloud Platform: Performance Benchmark and User Tutorial · Original source · Clinical and Translational Science 18(11):e70416, published 2025-11-18; PMC12627752 full-text XML
- Franzoso et al. 2025, Table S2 (runtime and costs per sample) and Table S1 (sample identifiers) · Original source · CTS-18-e70416-s002.xlsx (Table S2) inside the Europe PMC supplementary files bundle for PMC12627752
- Accelerating genomic workflows using NVIDIA Parabricks · Original source · BMC Bioinformatics 24:221, published 2023-05-31; PMC10230726 full-text XML
- Benchmarking accelerated next-generation sequencing analysis pipelines · Original source · Bioinformatics Advances 5(1):vbaf085, published 2025-05-15; PMC12092081 full-text XML
- Samarakoon et al. 2025, supplementary materials (Tables S1-S4, supplementary texts) · Original source · 'Publication-ready_Supplementary materials-20250404.pdf' inside vbaf085_supplementary_data.zip
- GeNePi: a graphics processing unit enhanced next-generation bioinformatics pipeline for whole-genome sequencing analysis · Original source · Briefings in Bioinformatics 27(1), published 2026-01-25; PMC12832024 full-text XML
- GeNePi: Supplementary Information · Original source · genepi_supplinfo_rev_bbag001.pdf from the Europe PMC supplementary files of PMC12832024
- A novel and accelerated method for integrated alignment and variant calling from short and long reads · Original source · Frontiers in Bioinformatics 5:1691056, published 2026-01-05; PMC12813096 full-text XML
- FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean · Original source · bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18
Technical metadata and extraction receipts
Stable ID: metadata-correction-7738c2859b8d725359fe
- field
- attributes.value
- previous value
- Three published execution benchmarks are linked, alongside the DRAGEN 4.2 developer total runtime for HG002 (1,838.5 seconds). All measure conventional variant-calling pipelines; none measures a genomic foundation model. None prints repeated runs, measured peak memory or accuracy next to runtime, so they do not show reproducibility, and hardware differs within each comparison, so differences mix software and hardware. On ten WGS samples run by a University of Oslo and Oslo University Hospital group with no declared competing interests (Samarakoon et al. 2025), read mapping plus HaplotypeCaller for NA12878 (47.55x) took 2,317.7 minutes on a CPU-only GATK pipeline, 123.45 minutes with Parabricks on four L4 GPUs, 78.12 on four A100 GPUs, 54.22 on eight H100 GPUs and 60.29 with DRAGEN v4.2. On the same H100 machine, Parabricks DeepVariant took 14.58 minutes for NA12878 with the version the pipeline used (printed 4.1.0-1; Methods say 4.3.0-1) and 16.05 minutes with the updated version (printed 4.3.0-1; Discussion says 4.3.1-1). On one HG002 30x genome on Google Cloud (O'Connell et al. 2023; the authors' employer funded the study and is an NVIDIA, AWS and Google alliance partner), HaplotypeCaller took 38.8 hours and cost 67.9 USD on a 32-vCPU machine and 0.59 hours and 17.5 USD on eight A100 GPUs, while the somatic caller LoFreq cost 8.1 USD on the CPU machine and 19 to 30.1 USD on 2 to 8 A100 GPUs. On Google Cloud machines of similar hourly cost (Franzoso et al. 2025, no declared conflicts), Sentieon DNASeq took 0:14:00 to 0:16:00 per exome at 0.82 to 1.03 USD and 3:00:00 to 3:50:00 per genome at 8.02 to 10.7 USD, and Parabricks Germline on one T4 GPU took 0:10:00 to 0:14:00 per exome at 0.71 to 0.93 USD and 4:06:00 to 4:40:00 per genome at 8.13 to 10.6 USD; costs exclude the Sentieon licence, and two genome runtime cells that cannot be genome runtimes are disputed and left out of these ranges. Each comparison is one run per sample on at most ten public samples, and costs are list prices at the time of each study.
- value
- Eight published execution benchmarks are linked, alongside the DRAGEN 4.2 developer total runtime for HG002 (1,838.5 seconds); two are withheld pending source concerns (see the end of this summary). Of the six active benchmarks, five measure conventional variant-calling pipelines and one measures inference of a genomic foundation model on microbial genomes, outside any diagnostic workflow. Only the foundation-model benchmark prints spread over repeated runs (five trials); none reruns a variant-calling pipeline to measure concordance of the calls, none prints variant-calling accuracy in the same table as runtime, and hardware differs within most comparisons, so differences mix software and hardware. On ten WGS samples run by a University of Oslo and Oslo University Hospital group with no declared competing interests (Samarakoon et al. 2025), read mapping plus HaplotypeCaller for NA12878 (47.55x) took 2,317.7 minutes on a CPU-only GATK pipeline, 123.45 minutes with Parabricks on four L4 GPUs, 78.12 on four A100 GPUs, 54.22 on eight H100 GPUs and 60.29 with DRAGEN v4.2. On the same H100 machine, Parabricks DeepVariant took 14.58 minutes for NA12878 with the version the pipeline used (printed 4.1.0-1; Methods say 4.3.0-1) and 16.05 minutes with the updated version (printed 4.3.0-1; Discussion says 4.3.1-1). On one HG002 30x genome on Google Cloud (O'Connell et al. 2023; the authors' employer funded the study and is an NVIDIA, AWS and Google alliance partner), HaplotypeCaller took 38.8 hours and cost 67.9 USD on a 32-vCPU machine and 0.59 hours and 17.5 USD on eight A100 GPUs, while the somatic caller LoFreq cost 8.1 USD on the CPU machine and 19 to 30.1 USD on 2 to 8 A100 GPUs. On Google Cloud machines of similar hourly cost (Franzoso et al. 2025, no declared conflicts), Sentieon DNASeq took 0:14:00 to 0:16:00 per exome at 0.82 to 1.03 USD and 3:00:00 to 3:50:00 per genome at 8.02 to 10.7 USD, and Parabricks Germline on one T4 GPU took 0:10:00 to 0:14:00 per exome at 0.71 to 0.93 USD and 4:06:00 to 4:40:00 per genome at 8.13 to 10.6 USD; costs exclude the Sentieon licence, and two genome runtime cells that cannot be genome runtimes are disputed and left out of these ranges. On human WGS at five read depths on the HPC nodes of a regional genome project (Marangoni et al. 2026; the GeNePi developers, no declared competing interests; one unnamed sample per depth), a 30x genome took 17h with the CPU GATK equivalent and 1h 36m, 42 m and 30m with Parabricks v4.3.0 on two V100, two A100 and four A100 GPUs, and a 100x genome 60h, 6h 30m, 2h 30m and 2h 06m. On one 120-vCPU Azure instance (Hu et al. 2026; all authors employed by Sentieon), DNAscope took 18.9 minutes for 30x Illumina, DNAscope LongRead 41.2 minutes for 30x PacBio HiFi and DNAscope Hybrid 85.8 minutes for 10x PacBio HiFi plus 35x Illumina, at 1.1, 2.5 and 5.2 USD on demand excluding the licence; each pipeline had different input data, so the three are not a ranking. For GenomeOcean on one H100 (Yu et al. 2026; two authors are at the DOE Joint Genome Institute, which develops GenomeOcean; GTDB bacterial genomes; mean and standard deviation over five trials), embedding extraction with the 4B model ran at 21,386 ± 126 input tokens per second with 72.14 GB peak GPU memory in BF16 and 22,401 ± 204 with 67.91 GB with FP8 weights, and generation ran at 1818.79 ± 7.95 generated tokens per second in BF16 and 2169.04 ± 24.30 with FP8 weights, activations and KV cache. No active linked result measures variant-calling accuracy against a truth set; the GenomeOcean ARI values score metagenomic binning against GTDB species labels. The two withheld studies are Zheng et al. 2026 (Clair3 developers; Clair3 and DeepVariant on HG003 long reads), whose supplementary runtimes, DeepVariant version and DeepVariant accuracy differ from its Results text and whose accuracy uses one GIAB sample's v4.2.1 confident regions, which leave out the hardest regions; and Aspidova et al. 2026 (ModernGENA and GENA-LM developers; four DNA language model families on one A100), whose printed tokens per second are far too low for its batch sizes. Apart from the five GenomeOcean trials, each comparison prints one value per sample and set-up on at most ten samples, and costs are list prices at the time of each study.
- source locator
- amp-source-dragen-supplementary-tables: result amp-result-dragen-hg002-phase4-runtime; model-execution-20261009-source-samarakoon2025-supplement: Supplementary Tables S1 and S3; model-execution-20261009-source-oconnell2023: Table 1; model-execution-20261009-source-franzoso2025-table-s2: Table S2; articles: Methods, Discussion, Funding and Conflicts of interest; model-execution-20261011-source-marangoni2026-genepi-supplement: Table S1; model-execution-20261011-source-marangoni2026-genepi: Competing interest; model-execution-20261011-source-hu2026-dnascope-hybrid: Table 2, Conflict of interest; model-execution-20261011-source-yu2026-fp8-genomeocean: Tables 3 and 4, author affiliations
- review
- method: automated-source-review; method note: Previous value copied byte for byte from the stored summary on origin/main; new value is the reviewer's approved summary text in docs/reviews/use-cases/diagnostic-genomics-model-execution-2026-10-12.md.; reviewer: claude; reviewer note: Claude (Opus 5.5) review agent reviewer-exec, independent of the collector uc-execution; no human review claimed; reviewed at: 2026-10-12T07:50:32Z; note: Summary text replaced as approved in the review; the previous value is kept here.