Protocol

Feng sQTL discrimination

Table 6 sQTL causal-versus-noncausal variant discrimination; Feng et al. 2025 source-reported protocol.

Overview

Table 6 sQTL causal-versus-noncausal variant discrimination; Feng et al. 2025 source-reported protocol.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

12 recorded evaluations, 24 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

12 evaluations · 24 results. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: AlphaGenome, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.715 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaGenome, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, AlphaGenome, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Source Results prose (Sec8) states AlphaGenome's QTL AUC as approximately 0.80, which differs from this exact Table 6 AUC cell (eQTL 0.8029, sQTL 0.7147); the table value is retained and the prose discrepancy is not silently resolved or transposed onto Cohen's d.
Configuration: AlphaGenome, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.787 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AlphaGenome, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, AlphaGenome, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Caduceus-Ph (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.567 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Caduceus-Ph sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Caduceus-Ph row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Caduceus-Ph (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.228 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Caduceus-Ph sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Caduceus-Ph row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Caduceus-Ph, long sequence (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.57 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Caduceus-Ph, long sequence sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Caduceus-Ph, long sequence row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Caduceus-Ph, long sequence (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.246 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Caduceus-Ph, long sequence sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Caduceus-Ph, long sequence row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: DNABERT-2 (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.58 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DNABERT-2 sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, DNABERT-2 row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: DNABERT-2 (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.282 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DNABERT-2 sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, DNABERT-2 row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Enformer, hidden states* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.666 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Enformer, hidden states* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Enformer, hidden states* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Enformer, hidden states* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.612 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Enformer, hidden states* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Enformer, hidden states* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Enformer, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.617 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Enformer, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Enformer, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Enformer, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.413 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Enformer, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Enformer, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: GROVER (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.474 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source GROVER sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, GROVER row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: GROVER (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	−0.0978 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source GROVER sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, GROVER row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: HyenaDNA-450K, long sequence (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.526 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source HyenaDNA-450K, long sequence sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, HyenaDNA-450K, long sequence row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: HyenaDNA-450K, long sequence (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination long-window dataset	0.0757 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source HyenaDNA-450K, long sequence sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, HyenaDNA-450K, long sequence row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: HyenaDNA (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.553 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source HyenaDNA sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, HyenaDNA row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: HyenaDNA (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.202 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source HyenaDNA sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, HyenaDNA row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: NT-v2 (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.505 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NT-v2 sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, NT-v2 row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: NT-v2 (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	−0.004 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NT-v2 sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, NT-v2 row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Sei, hidden states* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.653 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Sei, hidden states* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Sei, hidden states* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Sei, hidden states* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.553 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Sei, hidden states* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Sei, hidden states* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.
Configuration: Sei, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.628 sQTL AUC AUC · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Sei, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, AUC block, Sei, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md).
Configuration: Sei, output tracks* (sQTL discrimination)	Protocol: Feng sQTL discrimination Dataset: Feng sQTL discrimination short-window dataset	0.494 sQTL Cohen’s d Cohen's d · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source Sei, output tracks* sQTL discrimination evaluation Table 6, sQTL column, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 6, Cohen’s d block, Sei, output tracks* row, sQTL column (deterministic XML extraction; rowspans resolved per data/omics/amp-coverage-20261007/feng/review-notes.md). Signed effect size; no universal desirable sign/direction is stated by the source.

Source checking is not independent reproduction. Release 2026-10-07-1448159e6a81.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

External evaluations: 12

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Select a task-valid null control after reviewing inputs and metric

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Select an upstream conventional reference after reviewing the full protocol

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV · Model evaluation matrix · Source table · Release and checksums

Coverage is derived from release 2026-10-07-1448159e6a81. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

No runnable recipe has been reviewed for this protocol. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

Strengths, limitations and unresolved questions

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-07-1448159e6a81
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-10-07-1448159e6a81 · Record review: source checked

1 source records and release history

Benchmarking DNA foundation models for genomic and genetic tasks · Original source · PMC12663285.1

Download this release

Technical metadata and extraction receipts

Stable ID: amp-feng-20261007-protocol-qtl-sqtl

areas: dna-genomes
missing metadata: None recorded
scope note: Alternate-minus-reference embedding representations scored by a supervised random forest classifier. Three disjoint outer chromosome test groups (3/6/9/12/16/18/19/21; 2/5/11/14/17/20/22/X; 1/4/7/8/10/13/15), each held out in turn with fourfold inner-chromosome cross-validation for hyperparameter tuning. Borzoi-derived GTEx v8 whole-blood putative causal variants paired with matched noncausal variants by distance to functional site/gene expression. Pre-window-filter positive count for sQTL: 540 (not an established per-model scored denominator; post-filter counts are unreported and recorded as null, not zero). Short and long generated-window datasets are separate scored populations, never pooled. AlphaGenome uses the long dataset with a central input crop and separate output averaging.
task: sQTL causal-vs-noncausal variant discrimination (AUC, Cohen's d)
version: Table 6

Related records

protocol: AlphaGenome, output tracks* sQTL discrimination evaluation
protocol: Caduceus-Ph sQTL discrimination evaluation
protocol: Caduceus-Ph, long sequence sQTL discrimination evaluation
protocol: DNABERT-2 sQTL discrimination evaluation
protocol: Enformer, hidden states* sQTL discrimination evaluation
protocol: Enformer, output tracks* sQTL discrimination evaluation
protocol: GROVER sQTL discrimination evaluation
protocol: HyenaDNA sQTL discrimination evaluation
protocol: HyenaDNA-450K, long sequence sQTL discrimination evaluation
protocol: NT-v2 sQTL discrimination evaluation
protocol: Sei, hidden states* sQTL discrimination evaluation
protocol: Sei, output tracks* sQTL discrimination evaluation

Suggest a correction