rewirebio.iobenchmarks
← Browse all evidence claims
Evidence claim

Relevance of feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb to "Select regulatory variants and genes for functional follow-up"

Use case
Select regulatory variants and genes for functional follow-up
Assessed by
Feng GTEx whole-blood expression prediction, all genes, 6 kb input
Relevance
Proxy evidence
Endpoint
Average per-gene Pearson correlation and MSE between predicted and observed whole-blood expression across GTEx subjects, from frozen embeddings of each subject's phased TSS-centred sequence plus a per-gene random forest

Descriptive fact transcribed from the pinned source.

5 reviewed evaluations

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 2026-11-02-a7f03c9e55a5
Property and statementOriginal source and locationReview and provenance
Citation locations
1 values
  • source id
    evidence-expansion-dna-foundation-models-2025-5d8ca9bc
    locator
    Table 4; Methods, 'Gene expression prediction datasets' and 'Gene expression prediction benchmark'; Data availability
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.citation_locators

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Comparison group
feng-gtex-expression
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.comparison_group

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Comparison title
Feng et al. GTEx whole-blood expression prediction
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.comparison_title

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

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 rows only within this protocol.
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.constraints

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Endpoint
Average per-gene Pearson correlation and MSE between predicted and observed whole-blood expression across GTEx subjects, from frozen embeddings of each subject's phased TSS-centred sequence plus a per-gene random forest
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.endpoint

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Claimed field
links:assessed_by:feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.field

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Headline metric
pearson-correlation
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.headline_metric

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Limitations
8 values
  • Labels are covariate-corrected whole-blood expression of GTEx subjects, not allele-specific or endogenous perturbation effects, and not in the user's cell type.
  • Each model's representation is frozen; a random forest is fitted per gene on expression labels, so this measures how much expression variation the embedding plus a supervised head explains, not a label-free variant-effect score.
  • Individual-level GTEx data are protected (GTEx Protected Data Access application); the benchmark cannot be rerun from public data.
  • Transcribed from the paper; Feng et al. did not develop the models (independent_paper). No reproduction by Rewire and no human scientific review.
  • The authors call overall performance modest and say the findings should be considered preliminary indicators (Results, Gene expression prediction, last paragraph).
  • Subjects are split 75:25 within each gene; no loci, genes or studies are held out, which the use case's validation requirements ask for.
  • GROVER uses a 2,048 bp centre crop, so its input differs from the other four rows.
  • The ~6 kb TSS-centred window leaves out distal regulatory elements.
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.limitations

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Pinned records
21 values
  • record id
    evidence-expansion-dna-foundation-models-2025-5d8ca9bc
    fields
    • status
    • attributes.artifact_sha256
    sha256
    7996f91821054e875cfca234e2e3bca2b25a02af92475d0577cd3dd0a71a8559
  • record id
    evidence-reported-dna-foundation-models-2025-readme-md
    fields
    • status
    • attributes.artifact_sha256
    sha256
    58a64d048abd1fac81c829a001a39a4184741155de8eb41a05f8e8cf3edabea8
  • record id
    feng-dna-fm-20261013-eval-gtex-caduceus-ph
    fields
    • status
    • links
    • attributes.origin
    • attributes.comparison
    sha256
    3117284e6d39636fc7c99cfc0bcbb8aaa37a7fd1964bd6eafca8da4f6c7a484e
  • record id
    feng-dna-fm-20261013-eval-gtex-dnabert-2
    fields
    • status
    • links
    • attributes.origin
    • attributes.comparison
    sha256
    efed507098e5cf90867bf31905adb45e90f85e0daddb8999abef9f33007a76c7
  • record id
    feng-dna-fm-20261013-eval-gtex-grover
    fields
    • status
    • links
    • attributes.origin
    • attributes.comparison
    sha256
    9a3ef54b35a12caa6cf863f538a4fa2f84e6f53f75fc241c140a311045a09a50
  • record id
    feng-dna-fm-20261013-eval-gtex-hyenadna
    fields
    • status
    • links
    • attributes.origin
    • attributes.comparison
    sha256
    5daf7a2c5fa0c19b0a899be1f05406c26484e707bd949796d4226742245ff891
  • record id
    feng-dna-fm-20261013-eval-gtex-nt-v2
    fields
    • status
    • links
    • attributes.origin
    • attributes.comparison
    sha256
    5861c1513c6aecaab1e678884db5fa206429cc14a2dac944bc5192e415b84889
  • record id
    feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb
    fields
    • status
    • name
    • description
    • facets
    • source_ids
    • links
    • attributes
    sha256
    3207ea07e2c33c1c892c89d95951cce39cdc96b129d85c5662a00125f58fcc03
  • record id
    feng-dna-fm-20261013-result-gtex-caduceus-ph-mse
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    016c8546b24e653ce546235424bd78e8a6b103aadbe505abe3e9042cb277b154
  • record id
    feng-dna-fm-20261013-result-gtex-caduceus-ph-pearson
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    5d71f0f7b46d1d721e832c045f595813ac1e79c0189cda3a6734c58b2b8215c5
  • record id
    feng-dna-fm-20261013-result-gtex-dnabert-2-mse
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    c416be3c133c2ba77030784ceb06333c976f1e0424c7af57bef5874ec2fb37df
  • record id
    feng-dna-fm-20261013-result-gtex-dnabert-2-pearson
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    a71fb16cf36bed492896fb7096ab1c202d6fa1139e4cbef483bfb7106d366471
  • record id
    feng-dna-fm-20261013-result-gtex-grover-mse
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    bd804bb6fb64037f48e00b54c71fb7bc049adb6bb7f904b16a350d0a0a318361
  • record id
    feng-dna-fm-20261013-result-gtex-grover-pearson
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    e3175357ed868263b10f5dd026f615cfdb8353d2bbd52e249855db02c5e0334a
  • record id
    feng-dna-fm-20261013-result-gtex-hyenadna-mse
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    ae144b7b92d1ef5528a44c3a80c884705b30777145d7ed3c61020016aae8e889
  • record id
    feng-dna-fm-20261013-result-gtex-hyenadna-pearson
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    e81b68cffd09851c25acdb4246c7643a45b9cba5b95163f252ca1a09b116a69b
  • record id
    feng-dna-fm-20261013-result-gtex-nt-v2-mse
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    c416be3c133c2ba77030784ceb06333c976f1e0424c7af57bef5874ec2fb37df
  • record id
    feng-dna-fm-20261013-result-gtex-nt-v2-pearson
    fields
    • status
    • attributes.metric
    • attributes.metric_qualifier
    • attributes.printed_value
    • attributes.numeric_value
    sha256
    e81b68cffd09851c25acdb4246c7643a45b9cba5b95163f252ca1a09b116a69b
  • record id
    feng-dna-fm-20261013-source-gtex-hyena-script
    fields
    • status
    • attributes.artifact_sha256
    sha256
    14826c6e46f84645a24696ddc49fddd5506047571f168932b6e3d7a4ef296f43
  • record id
    feng-dna-fm-20261013-source-supplementary-information
    fields
    • status
    • attributes.artifact_sha256
    sha256
    b4b55ba9044d429e526d69c577581951682d9a8668b3e44820badecf3448e42f
  • 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		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.pins

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Rationale
Proxy: tests whether a model's sequence representation captures inter-individual cis-sequence effects on expression, which bears on allele-effect prediction for a locus. It does not score individual variants, link elements to genes or measure perturbation outcomes, which the use case needs.
Context-only references		Benchmarking DNA foundation models for genomic and genetic tasks

Original source ↗

evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4

Version: PMC12663285.1
Retrieved: 2026-09-17T07:54:51.004119+00:00

not individually reviewed

No individual claim review recorded

Audit details

Field: attributes.rationale

Source artifact SHA-256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

How to read the evidence

Sources and history

Release 2026-11-02-a7f03c9e55a5 · Record review: source checked

1 source record and release historyDownload this release (gzip)
Technical metadata and extraction receipts

Stable ID: use-case-mapping-feng-dna-fm-20261013-gtex-wb-expression-6kb

field
links:assessed_by:feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb
value
feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb
relevance
proxy
endpoint
Average per-gene Pearson correlation and MSE between predicted and observed whole-blood expression across GTEx subjects, from frozen embeddings of each subject's phased TSS-centred sequence plus a per-gene random forest
rationale
Proxy: tests whether a model's sequence representation captures inter-individual cis-sequence effects on expression, which bears on allele-effect prediction for a locus. It does not score individual variants, link elements to genes or measure perturbation outcomes, which the use case needs.
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 rows only within this protocol.
limitations
Labels are covariate-corrected whole-blood expression of GTEx subjects, not allele-specific or endogenous perturbation effects, and not in the user's cell type.; Each model's representation is frozen; a random forest is fitted per gene on expression labels, so this measures how much expression variation the embedding plus a supervised head explains, not a label-free variant-effect score.; Individual-level GTEx data are protected (GTEx Protected Data Access application); the benchmark cannot be rerun from public data.; Transcribed from the paper; Feng et al. did not develop the models (independent_paper). No reproduction by Rewire and no human scientific review.; The authors call overall performance modest and say the findings should be considered preliminary indicators (Results, Gene expression prediction, last paragraph).; Subjects are split 75:25 within each gene; no loci, genes or studies are held out, which the use case's validation requirements ask for.; GROVER uses a 2,048 bp centre crop, so its input differs from the other four rows.; The ~6 kb TSS-centred window leaves out distal regulatory elements.
citation locators
source id: evidence-expansion-dna-foundation-models-2025-5d8ca9bc; locator: Table 4; Methods, 'Gene expression prediction datasets' and 'Gene expression prediction benchmark'; Data availability
source locator
evidence-expansion-dna-foundation-models-2025-5d8ca9bc: Table 4
headline metric
pearson-correlation
comparison group
feng-gtex-expression
comparison title
Feng et al. GTEx whole-blood expression prediction
stratum label
All 21,004 genes, ~6 kb input
stratum order
1
reason
Add a relevance judgement for the Feng et al. gene-expression protocol (rewire-benchmark-data#19).
reviewed evaluations
feng-dna-fm-20261013-eval-gtex-caduceus-ph; feng-dna-fm-20261013-eval-gtex-dnabert-2; feng-dna-fm-20261013-eval-gtex-grover; feng-dna-fm-20261013-eval-gtex-hyenadna; feng-dna-fm-20261013-eval-gtex-nt-v2
review
method: ai-assisted-source-review; scope-review; method note: Checked the endpoint against the use case's inputs, output and setting in data/entities/use-cases.jsonl; checked each limitation against Methods, Results, Data availability and Table 4; checked every evaluation on the protocol.; reviewer: claude; reviewer note: Independent Claude (Opus 5.5) review agent, separate from the extraction agent feng-19 that built this batch. Automated review only; no human review claimed. Transcription, not reproduction.; reviewed at: 2026-10-13T14:40:00Z; artifact sha256: 5d8ca9bcf88cc1b38ad667906a2e4699b1aefa6d31c6f49259784930353f3202; retrieval url: https://www.ebi.ac.uk/europepmc/webservices/rest/PMC12663285/fullTextXML; note: Judgement written by the extraction agent. Independent review: proxy is justified because cross-individual expression prediction bears only on allele-effect prediction, which the use case treats as a separate comparison; it does not score variants or link elements to genes.
pins
record id: evidence-expansion-dna-foundation-models-2025-5d8ca9bc; fields: status; attributes.artifact_sha256; sha256: 7996f91821054e875cfca234e2e3bca2b25a02af92475d0577cd3dd0a71a8559; record id: evidence-reported-dna-foundation-models-2025-readme-md; fields: status; attributes.artifact_sha256; sha256: 58a64d048abd1fac81c829a001a39a4184741155de8eb41a05f8e8cf3edabea8; record id: feng-dna-fm-20261013-eval-gtex-caduceus-ph; fields: status; links; attributes.origin; attributes.comparison; sha256: 3117284e6d39636fc7c99cfc0bcbb8aaa37a7fd1964bd6eafca8da4f6c7a484e; record id: feng-dna-fm-20261013-eval-gtex-dnabert-2; fields: status; links; attributes.origin; attributes.comparison; sha256: efed507098e5cf90867bf31905adb45e90f85e0daddb8999abef9f33007a76c7; record id: feng-dna-fm-20261013-eval-gtex-grover; fields: status; links; attributes.origin; attributes.comparison; sha256: 9a3ef54b35a12caa6cf863f538a4fa2f84e6f53f75fc241c140a311045a09a50; record id: feng-dna-fm-20261013-eval-gtex-hyenadna; fields: status; links; attributes.origin; attributes.comparison; sha256: 5daf7a2c5fa0c19b0a899be1f05406c26484e707bd949796d4226742245ff891; record id: feng-dna-fm-20261013-eval-gtex-nt-v2; fields: status; links; attributes.origin; attributes.comparison; sha256: 5861c1513c6aecaab1e678884db5fa206429cc14a2dac944bc5192e415b84889; record id: feng-dna-fm-20261013-protocol-gtex-wb-expression-6kb; fields: status; name; description; facets; source_ids; links; attributes; sha256: 3207ea07e2c33c1c892c89d95951cce39cdc96b129d85c5662a00125f58fcc03; record id: feng-dna-fm-20261013-result-gtex-caduceus-ph-mse; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: 016c8546b24e653ce546235424bd78e8a6b103aadbe505abe3e9042cb277b154; record id: feng-dna-fm-20261013-result-gtex-caduceus-ph-pearson; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: 5d71f0f7b46d1d721e832c045f595813ac1e79c0189cda3a6734c58b2b8215c5; record id: feng-dna-fm-20261013-result-gtex-dnabert-2-mse; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: c416be3c133c2ba77030784ceb06333c976f1e0424c7af57bef5874ec2fb37df; record id: feng-dna-fm-20261013-result-gtex-dnabert-2-pearson; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: a71fb16cf36bed492896fb7096ab1c202d6fa1139e4cbef483bfb7106d366471; record id: feng-dna-fm-20261013-result-gtex-grover-mse; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: bd804bb6fb64037f48e00b54c71fb7bc049adb6bb7f904b16a350d0a0a318361; record id: feng-dna-fm-20261013-result-gtex-grover-pearson; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: e3175357ed868263b10f5dd026f615cfdb8353d2bbd52e249855db02c5e0334a; record id: feng-dna-fm-20261013-result-gtex-hyenadna-mse; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: ae144b7b92d1ef5528a44c3a80c884705b30777145d7ed3c61020016aae8e889; record id: feng-dna-fm-20261013-result-gtex-hyenadna-pearson; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: e81b68cffd09851c25acdb4246c7643a45b9cba5b95163f252ca1a09b116a69b; record id: feng-dna-fm-20261013-result-gtex-nt-v2-mse; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: c416be3c133c2ba77030784ceb06333c976f1e0424c7af57bef5874ec2fb37df; record id: feng-dna-fm-20261013-result-gtex-nt-v2-pearson; fields: status; attributes.metric; attributes.metric_qualifier; attributes.printed_value; attributes.numeric_value; sha256: e81b68cffd09851c25acdb4246c7643a45b9cba5b95163f252ca1a09b116a69b; record id: feng-dna-fm-20261013-source-gtex-hyena-script; fields: status; attributes.artifact_sha256; sha256: 14826c6e46f84645a24696ddc49fddd5506047571f168932b6e3d7a4ef296f43; record id: feng-dna-fm-20261013-source-supplementary-information; fields: status; attributes.artifact_sha256; sha256: b4b55ba9044d429e526d69c577581951682d9a8668b3e44820badecf3448e42f; 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

Suggest a correction