Strengths and considerations
No source-reviewed explanatory claims are recorded here yet.
benchmark · protocol
CAGE track prediction at 32 bp (AlphaGenome paper)
Can reference sequence predict held-out CAGE signal?
AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-trainingAt a glance
Explanatory profile: limited source coverage · Automated source review, 2026-09-17. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Dataset and biological context | Original Borzoi TFRecords, retaining their scaling and 32-bp target bins.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Split | Held-out human genomic intervals under the relevant fold. General training follows Borzoi folds, extending intervals to the AlphaGenome context and removing validation/test windows overlapping training after extension; use the comparison-specific split below rather than the variant chromosome split. This comparison uses Borzoi fold-1.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Allowed inputs and adaptation | Reference DNA and the dedicated Borzoi-matched AlphaGenome output head; compare published Borzoi fold-1 predictions.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Metrics as reported | pearsonrAlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Aggregation | Pearson correlation; group summaries derive from individual track correlations.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Uncertainty | Not reported for these summary-table scores. · Not reported in inspected sourcesAlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training |
| Entity type | protocol |
| Organisms | Not extracted or verified for this record. |
| Assays | Not extracted or verified for this record. |
| Baselines | Not extracted or verified for this record. |
How it works
Conceptual summary of the cited procedure; model-specific conditions are given below.
Read the diagram as text
- Borzoi fold-1 holdout
- Predict matched 32-bp tracks
- Align target scale and bins
- Summarize Pearson correlation
What is tested
Can reference sequence predict held-out CAGE signal?
AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-trainingProcedure
Align predicted and experimental values on the same 32-bp tracks and test intervals.
AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-trainingTested models and results
Release 2026-09-17-a757f4af4277 · 2 evaluations · 2 metric rows. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| AlphaGenome Borzoi-fold-1 model fine-tuned with 32 bp Borzoi-matching head: CAGE track prediction at 32 bp Model: AlphaGenome Borzoi-fold-1 model fine-tuned with 32 bp Borzoi-matching head · Benchmark: CAGE track prediction at 32 bp (AlphaGenome paper) · Dataset: CAGE track prediction at 32 bp: evaluated data subset Align predicted and experimental values on the same 32-bp tracks and test intervals. Author-reported evaluation · Evaluation metadata: needs review | ||
| 0.74 pearsonr Unit: correlation · Direction: higher Aggregation: Pearson correlation; group summaries derive from individual track correlations. | Uncertainty: unreported Scored: Not reported · Eligible: Not reported | source checkedAlphaGenome Nature 2026 supplementary comparison tables · 'Suppl Table 3 Track performance'!K18 Source checking is not independent reproduction. |
| Borzoi (paper Table 3): CAGE track prediction at 32 bp Model: Borzoi (paper Table 3) · Benchmark: CAGE track prediction at 32 bp (AlphaGenome paper) · Dataset: CAGE track prediction at 32 bp: evaluated data subset Align predicted and experimental values on the same 32-bp tracks and test intervals. Author-reported evaluation · Evaluation metadata: needs review | ||
| 0.713 pearsonr Unit: correlation · Direction: higher Aggregation: Pearson correlation; group summaries derive from individual track correlations. | Uncertainty: unreported Scored: Not reported · Eligible: Not reported | source checkedAlphaGenome Nature 2026 supplementary comparison tables · 'Suppl Table 3 Track performance'!J18 Source checking is not independent reproduction. |
Strengths and limitations
Limitations and conditions
- This is a fine-tuned comparison head, not the generic distilled variant model. Scaling and resolution differ from the separate base-resolution RNA endpoint.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training
- This is an author-reported protocol, not a rewire rerun. Different datasets, processing and adaptations cannot support an unrestricted leaderboard.AlphaGenome Nature 2026 supplementary comparison tables; AlphaGenome Nature 2026 supplementary methods; alphagenome: Journal full-text XML · 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training
Profile review details
Primary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction.
Stable record: alphagenome-2026-t3-protocol-10Evidence table
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.
42 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual summary of the cited procedure; model-specific conditions are given below. Individual claims | alphagenome: Journal full-text XML 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram caption Conceptual summary of the cited procedure; model-specific conditions are given below. Individual claims | AlphaGenome Nature 2026 supplementary methods 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Supplement to Nature version of record, 28 January 2026; content hash pinned | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact publisher PDF bytes Format: original_pdf |
| Diagram caption Conceptual summary of the cited procedure; model-specific conditions are given below. Individual claims | AlphaGenome Nature 2026 supplementary comparison tables 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Nature version of record, 28 January 2026 | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact retrieved original artifact bytes Format: xlsx |
| Diagram steps ["Borzoi fold-1 holdout","Predict matched 32-bp tracks","Align target scale and bins","Summarize Pearson correlation"] Individual claims | alphagenome: Journal full-text XML 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram steps ["Borzoi fold-1 holdout","Predict matched 32-bp tracks","Align target scale and bins","Summarize Pearson correlation"] Individual claims | AlphaGenome Nature 2026 supplementary methods 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Supplement to Nature version of record, 28 January 2026; content hash pinned | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact publisher PDF bytes Format: original_pdf |
| Diagram steps ["Borzoi fold-1 holdout","Predict matched 32-bp tracks","Align target scale and bins","Summarize Pearson correlation"] Individual claims | AlphaGenome Nature 2026 supplementary comparison tables 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Nature version of record, 28 January 2026 | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact retrieved original artifact bytes Format: xlsx |
| Diagram title CAGE track prediction at 32 bp: evaluation procedure Individual claims | alphagenome: Journal full-text XML 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram title CAGE track prediction at 32 bp: evaluation procedure Individual claims | AlphaGenome Nature 2026 supplementary methods 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Supplement to Nature version of record, 28 January 2026; content hash pinned | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact publisher PDF bytes Format: original_pdf |
| Diagram title CAGE track prediction at 32 bp: evaluation procedure Individual claims | AlphaGenome Nature 2026 supplementary comparison tables 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Nature version of record, 28 January 2026 | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of exact retrieved original artifact bytes Format: xlsx |
| Dataset and biological context Original Borzoi TFRecords, retaining their scaling and 32-bp target bins. Individual claims | alphagenome: Journal full-text XML 'Suppl Table 3 Track performance'!A18:N18; Supplementary Methods p.24, Benchmarking Against Borzoi Track Predictions; methods: pp.21–22, Genome Track Evaluation / Gene Expression Correlation; methods: p.24, Benchmarking Against Borzoi Track Predictions; paper: Extended Data Fig.3c,d; tables: Suppl Table 3 Track performance; evaluation index 10; sheet rows 18; methods: p.19, Training / Pre-training Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied | source checked automated source review · 2026-09-17 Audit detailsPrimary-source transcription and separate automated source review. No human sign-off or independent experimental reproduction. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
Sources and history
Release 2026-09-17-a757f4af4277 · Record review: needs review
- AlphaGenome Nature 2026 supplementary comparison tables · Original source · Nature version of record, 28 January 2026
- AlphaGenome Nature 2026 supplementary methods · Original source · Supplement to Nature version of record, 28 January 2026; content hash pinned
- alphagenome: Journal full-text XML · Original source · Retrieved page snapshot; no immutable publisher revision supplied
Technical metadata and extraction receipts
Stable ID: alphagenome-2026-t3-protocol-10
- areas
- dna-genomes
- entity level
- protocol
- version
- Nature version of record, 28 January 2026
- source evaluation index
- 10
- source table
- 3
- reference levels
- metric: pearsonr; printed value: 0; numeric value: 0; source locator: Suppl Table 3 Track performance!H18; note: Source reference quantity for relative-performance calculation, not a measured baseline run.