Sei, hidden states* (pathogenic/common SNP classification)
Sei, hidden states (pathogenic/common SNP classification): Sei, hidden states* (pathogenic/common SNP classification); Actual DNA input consumed by the model: 4096 bp (a crop of the dataset's generated window where smaller); Alternate-minus-reference frozen representation.
Overview
An explanatory profile has not yet been reviewed for this record.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluations and results
2 evaluations · 5 results. Different protocols are not a single leaderboard.
Protocol: Feng pathogenic-versus-common SNP classification · Dataset: Feng pathogenic/common SNP short-window dataset
Sorted by AUROC (task: pathogenic vs common snp; outer test group 1 only (chromosomes 3, 6, 9, 12, 16, 18, 19, 21), not averaged) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.
| Tested configuration | AUROC (task: pathogenic vs common snp; outer test group 1 only (chromosomes 3, 6, 9, 12, 16, 18, 19, 21), not averaged) | AUROC (task: pathogenic vs common snp; outer test group 2 only (chromosomes 2, 5, 11, 14, 17, 20, 22, X), not averaged) | AUROC (task: pathogenic vs common snp; outer test group 3 only (chromosomes 1, 4, 7, 8, 10, 13, 15), not averaged) | AUROC (task: pathogenic vs common snp) | Cohen's d (task: pathogenic vs common snp) |
|---|---|---|---|---|---|
| Sei, hidden states* (pathogenic/common SNP classification)Evaluation | 0.686 | 0.647 | 0.646 | Not reported | Not reported |
| Sei, hidden states* (pathogenic/common SNP classification)Evaluation | Not reported | Not reported | Not reported | 0.66 | 0.557 |
All 5 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: Sei, hidden states* (pathogenic/common SNP classification) | Protocol: Feng pathogenic-versus-common SNP classification Dataset: Feng pathogenic/common SNP short-window dataset | 0.66 AUROC (task: pathogenic vs common snp) unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceSei, hidden states* pathogenic/common SNP evaluation Table 5, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 5, Sei, hidden states* row, AUC column (deterministic XML extraction) |
| Configuration: Sei, hidden states* (pathogenic/common SNP classification) | Protocol: Feng pathogenic-versus-common SNP classification Dataset: Feng pathogenic/common SNP short-window dataset | 0.557 Cohen's d (task: pathogenic vs common snp) unitless · unknown Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceSei, hidden states* pathogenic/common SNP evaluation Table 5, frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks · Table 5, Sei, hidden states* row, Cohen's d column (deterministic XML extraction). Signed effect size; the source does not state a universal desirable sign/direction, so direction is recorded as unknown rather than assumed higher-is-better. |
| Configuration: Sei, hidden states* (pathogenic/common SNP classification) | Protocol: Feng pathogenic-versus-common SNP classification Dataset: Feng pathogenic/common SNP short-window dataset | 0.686 AUROC (task: pathogenic vs common snp; outer test group 1 only (chromosomes 3, 6, 9, 12, 16, 18, 19, 21), not averaged) unitless · higher Uncertainty: Not reported by the source: Supplementary Data 4 prints one AUC per test group, with no interval. Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceSei, hidden states Pathogenic variants vs common per-test-group evaluation Table 5 set-up, AUC reported separately for each outer chromosome test group (Supplementary Data 4); frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks: Supplementary Data 1-6 · Supplementary Data 4 (workbook sheet 'Data 4'), block 'Pathogenic variants vs common', row 'Sei, hidden states', column 'Test Group 1 AUC' (cell B12). Test group 1 is chromosomes 3, 6, 9, 12, 16, 18, 19, 21 (Methods, 'Variant effect quantification benchmark'). |
| Configuration: Sei, hidden states* (pathogenic/common SNP classification) | Protocol: Feng pathogenic-versus-common SNP classification Dataset: Feng pathogenic/common SNP short-window dataset | 0.647 AUROC (task: pathogenic vs common snp; outer test group 2 only (chromosomes 2, 5, 11, 14, 17, 20, 22, X), not averaged) unitless · higher Uncertainty: Not reported by the source: Supplementary Data 4 prints one AUC per test group, with no interval. Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceSei, hidden states Pathogenic variants vs common per-test-group evaluation Table 5 set-up, AUC reported separately for each outer chromosome test group (Supplementary Data 4); frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks: Supplementary Data 1-6 · Supplementary Data 4 (workbook sheet 'Data 4'), block 'Pathogenic variants vs common', row 'Sei, hidden states', column 'Test Group 2 AUC' (cell C12). Test group 2 is chromosomes 2, 5, 11, 14, 17, 20, 22, X (Methods, 'Variant effect quantification benchmark'). |
| Configuration: Sei, hidden states* (pathogenic/common SNP classification) | Protocol: Feng pathogenic-versus-common SNP classification Dataset: Feng pathogenic/common SNP short-window dataset | 0.646 AUROC (task: pathogenic vs common snp; outer test group 3 only (chromosomes 1, 4, 7, 8, 10, 13, 15), not averaged) unitless · higher Uncertainty: Not reported by the source: Supplementary Data 4 prints one AUC per test group, with no interval. Coverage: Not reported scored / Not reported eligible | Independent external evaluation · Source checkedMethods, coverage and sourceSei, hidden states Pathogenic variants vs common per-test-group evaluation Table 5 set-up, AUC reported separately for each outer chromosome test group (Supplementary Data 4); frozen embeddings + random forest classifier Aggregation: Not reported Benchmarking DNA foundation models for genomic and genetic tasks: Supplementary Data 1-6 · Supplementary Data 4 (workbook sheet 'Data 4'), block 'Pathogenic variants vs common', row 'Sei, hidden states', column 'Test Group 3 AUC' (cell D12). Test group 3 is chromosomes 1, 4, 7, 8, 10, 13, 15 (Methods, 'Variant effect quantification benchmark'). |
Source checking is not independent reproduction. Release 2026-11-02-a7f03c9e55a5.
Use this configuration
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.
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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
Facets: method types1 values
| Benchmarking DNA foundation models for genomic and genetic tasks The store's Enformer and Sei model records (regulatory-variant-20261009-model-enformer, regulatory-variant-20261009-model-sei) are supervised_machine_learning Version: PMC12663285.1 | conflicting claim ["automated-source-review"] · 2026-10-12 Audit detailsCorrect the method type from conventional_pipeline to supervised_machine_learning (use-case baseline review 2026-10-12, table 2). The review did not read the Feng source. These configurations have no method link. catalog-model-alphagenome is tagged foundation_model; Feng prints AlphaGenome with an asterisk as a non-DNA-foundation comparator. Evidence from docs/reviews/use-cases/baselines-2026-10-12.md. Independent review 2026-10-12: approved. See docs/reviews/2026-10-12-method-type-corrections.md. Conflicting source claim: Field: Claim: metadata-correction-tags-20261012-3df2db5190e747e8e59a Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-11-02-a7f03c9e55a5 · Record review: source checked
1 source record and release history
Download this release (gzip)Technical metadata and extraction receipts
Stable ID: amp-feng-20261007-config-variant-sei-hidden-states
- method types
- supervised_machine_learning
- actual input bp
- 4096
- foundation model eligible
- false
- head
- Supervised random forest fitted separately for this task
- missing metadata
- head revision: reason: unextracted; output aggregation bp: reason: unextracted; version: reason: unreported; note: Printed model name and input representation only; immutable checkpoint/head revision is unreported in the inspected primary text.
- reported name
- Sei, hidden states*
- representation
- Alternate-minus-reference frozen representation
- task
- pathogenic/common SNP classification
- provenance note
- Sei, hidden states* (pathogenic/common SNP classification); Feng et al. 2025 source-reported, task-scoped configuration. Printed in the source with an asterisk as a non-DNA-foundation specialised genomic comparator, not one of the paper's own DNA foundation models. Actual DNA input consumed by the model: 4096 bp (a crop of the dataset's generated window where smaller).