Evidence claim

Relevance of GFMBench-API TraitGym Complex, zs-ar-llr to "Select regulatory variants and genes for functional follow-up"

Use case: Select regulatory variants and genes for functional follow-up
Assessed by: GFMBench-API TraitGym Complex, zs-ar-llr
Relevance: Proxy evidence
Endpoint: Zero-shot AUROC and AUPRC of autoregressive log-likelihood ratio for Evo2 (1B, NVIDIA BioNeMo, FP16) on TraitGym Complex

GFMBench-API zero-shot autoregressive log-likelihood ratio scores on TraitGym Complex as evidence for choosing noncoding variants to edit.

1 reviewed evaluation

Evo2 on GFMBench-API TraitGym Complex (zs-ar-llr)

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.

19 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance
Citation locations 1 values source id lit-genomics-20261011-gfmbench-source-larey2026 locator Table 5, task 'TraitGym Complex'; appendix task description Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.citation_locators` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison group gfmbench-traitgym Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.comparison_group` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison title GFMBench-API zero-shot scores on TraitGym Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.comparison_title` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Constraints 3 values Inspect every linked evaluation's source locator before citing a result. Do not combine this mapping's evaluations with any other protocol's results. Compare only with other rows of the same GFMBench-API table; TraitGym's own paper uses different context lengths and scoring. Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.constraints` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Endpoint Zero-shot AUROC and AUPRC of autoregressive log-likelihood ratio for Evo2 (1B, NVIDIA BioNeMo, FP16) on TraitGym Complex Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.endpoint` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Claimed field links:assessed_by:lit-genomics-20261011-gfmbench-protocol-traitgym-complex-zs-ar-llr Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.field` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Headline metric auroc Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.headline_metric` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Limitations 3 values Preprint, not peer reviewed. Complex-trait AUROCs are all close to 0.5; differences between models are small and no uncertainty is printed. Fine-mapping labels, not experimental allele effects. Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.limitations` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Pinned records 6 values record id lit-genomics-20261011-gfmbench-eval-evo2-zero-shot-traitgym-complex-zs-ar-llr fields status links attributes.origin attributes.comparison sha256 10adb0cc8631c7b3b5661903fcf5e042323b60fd8ccf8262ae13c223b6b9a744 record id lit-genomics-20261011-gfmbench-protocol-traitgym-complex-zs-ar-llr fields status name description facets source_ids links attributes sha256 64cb3db3f75a0149629b7ae6326354c1085b80d3d99fd4ba609cc9d0affbc466 record id lit-genomics-20261011-gfmbench-result-evo2-zero-shot-traitgym-complex-zs-ar-llr-mean-llr-prc fields status attributes.metric attributes.metric_qualifier attributes.printed_value attributes.numeric_value sha256 a4a20d7e81c183d77da8028aa48cf172ca1453b8eeb662b9697ada68080d665c record id lit-genomics-20261011-gfmbench-result-evo2-zero-shot-traitgym-complex-zs-ar-llr-mean-llr-roc fields status attributes.metric attributes.metric_qualifier attributes.printed_value attributes.numeric_value sha256 8600714bc673a70c5b43f0a0005f0ca9d9793ccb2dc91f2b93f4659f6fb05609 record id lit-genomics-20261011-gfmbench-source-larey2026 fields status attributes.artifact_sha256 sha256 a461e55e4db9e70efca0dd40a686ae239e2e254757de72623099c721fd6eb0fc record id use-case-regulatory-variant-gene-follow-up fields status name description facets attributes.question attributes.decision attributes.inputs attributes.output attributes.exclusions attributes.setting attributes.clinical_scope attributes.intended_users sha256 0071547e01ccaec30238aea9e4882dd92e6f51be9c967c7fb88bc61d2788b870 Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.pins` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Rationale Proxy, not direct: the protocol ranks candidate noncoding variants by zero-shot model scores, which matches the first step of choosing allele edits, but it gives no element-gene link and no effect direction, and it is scored on the GFMBench-API re-implementation of TraitGym rather than on perturbation data. Context-only references	GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models Original source ↗ lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description Version: bioRxiv preprint version 2, posted 2026-08-11 Retrieved: 2026-10-11T00:44:58Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.rationale` Source artifact SHA-256: `df9934c88ee245b1e19e155174169d6b6da57ae07c1a04bbea7115423facdf72` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

1 source record and release history

GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models · Original source · bioRxiv preprint version 2, posted 2026-08-11

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: use-case-mapping-lit-genomics-20261011-gfmbench-traitgym-complex-zs-ar-llr

field: links:assessed_by:lit-genomics-20261011-gfmbench-protocol-traitgym-complex-zs-ar-llr
value: lit-genomics-20261011-gfmbench-protocol-traitgym-complex-zs-ar-llr
relevance: proxy
endpoint: Zero-shot AUROC and AUPRC of autoregressive log-likelihood ratio for Evo2 (1B, NVIDIA BioNeMo, FP16) on TraitGym Complex
rationale: Proxy, not direct: the protocol ranks candidate noncoding variants by zero-shot model scores, which matches the first step of choosing allele edits, but it gives no element-gene link and no effect direction, and it is scored on the GFMBench-API re-implementation of TraitGym rather than on perturbation data.
constraints: Inspect every linked evaluation's source locator before citing a result.; Do not combine this mapping's evaluations with any other protocol's results.; Compare only with other rows of the same GFMBench-API table; TraitGym's own paper uses different context lengths and scoring.
limitations: Preprint, not peer reviewed.; Complex-trait AUROCs are all close to 0.5; differences between models are small and no uncertainty is printed.; Fine-mapping labels, not experimental allele effects.
citation locators: source id: lit-genomics-20261011-gfmbench-source-larey2026; locator: Table 5, task 'TraitGym Complex'; appendix task description
source locator: lit-genomics-20261011-gfmbench-source-larey2026: Table 5, task 'TraitGym Complex'; appendix task description
headline metric: auroc
reason: Add a relevance judgement for a genomic foundation model benchmark protocol from the 2026-10-11 genomics suite pass, reviewed in docs/reviews/2026-10-11-genomics-gfm-suites.md.
comparison group: gfmbench-traitgym
comparison title: GFMBench-API zero-shot scores on TraitGym
stratum label: TraitGym Complex, autoregressive log-likelihood ratio
stratum order: 6
reviewed evaluations: lit-genomics-20261011-gfmbench-eval-evo2-zero-shot-traitgym-complex-zs-ar-llr
review: method: ai-assisted-source-review; scope-review; method note: Reviewer checked the endpoint against the use case's inputs and output, the proxy rating and the limitations; every evaluation on the protocol had its values checked in the batch review. Corrections listed in review.md were applied by the extractor.; reviewer: claude; reviewer note: Claude review agent reviewer-gen-3, independent of the extractor lit-genomics; no human review claimed; reviewed at: 2026-10-11T14:40:00Z; note: Automated review; not human scientific review.
pins: record id: lit-genomics-20261011-gfmbench-eval-evo2-zero-shot-traitgym-complex-zs-ar-llr; fields: status; links; attributes.origin; attributes.comparison; sha256: 10adb0cc8631c7b3b5661903fcf5e042323b60fd8ccf8262ae13c223b6b9a744; record id: lit-genomics-20261011-gfmbench-protocol-traitgym-complex-zs-ar-llr; fields: status; name; description; facets; source_ids; links; attributes; sha256: 64cb3db3f75a0149629b7ae6326354c1085b80d3d99fd4ba609cc9d0affbc466; record id: lit-genomics-20261011-gfmbench-result-evo2-zero-shot-traitgym-complex-zs-ar-llr-mean-llr-prc; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: a4a20d7e81c183d77da8028aa48cf172ca1453b8eeb662b9697ada68080d665c; record id: lit-genomics-20261011-gfmbench-result-evo2-zero-shot-traitgym-complex-zs-ar-llr-mean-llr-roc; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: 8600714bc673a70c5b43f0a0005f0ca9d9793ccb2dc91f2b93f4659f6fb05609; record id: lit-genomics-20261011-gfmbench-source-larey2026; fields: status; attributes.artifact_sha256; sha256: a461e55e4db9e70efca0dd40a686ae239e2e254757de72623099c721fd6eb0fc; record id: use-case-regulatory-variant-gene-follow-up; fields: status; name; description; facets; attributes.question; attributes.decision; attributes.inputs; attributes.output; attributes.exclusions; attributes.setting; attributes.clinical_scope; attributes.intended_users; sha256: 0071547e01ccaec30238aea9e4882dd92e6f51be9c967c7fb88bc61d2788b870

Related records

subject: Select regulatory variants and genes for functional follow-up

Suggest a correction