MCC (mean over three task categories of median cross-validation MCC) of NT 500M Randomly Initialized, trained from scratch (NT benchmark) on NT benchmark, all 18 tasks
0.355 MCC (mean over three task categories of median cross-validation MCC)
Methods
- Tested configuration
- NT 500M Randomly Initialized, trained from scratch (NT benchmark)
- Protocol
- NT benchmark normalised summed performance (Supplementary Table 9)
- Dataset
- NT benchmark, all 18 tasks
- Procedure
- NT benchmark normalised summed performance (Supplementary Table 9)
- Evaluation
- NT 500M Randomly Initialized (Training from scratch) NT benchmark normalised summed performance
- Coverage
- Not reported scored / Not reported eligible
- Uncertainty
- Not reported by the source
- Evidence
- Author-reported evaluation · source checkedNucleotide Transformer: Supplementary Information (Supplementary Tables 1-12) · Supplementary Table 9, row 'NT 500M Randomly Initialized', approach 'Training from scratch', column 'Average performance'
A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: source checked. Source checked does not mean independently reproduced.
Reproduction
- Split
- Tenfold cross-validation
- Adaptation
- trained from scratch
- Scoring implementation
- MCC
No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.
Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.
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 |
|---|---|---|
| Reported result 0.355 Individual claims | Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12) Supplementary Table 9, row 'NT 500M Randomly Initialized', approach 'Training from scratch', column 'Average performance' Version: Nature Methods 22, 287-297 (published online 2024-11-28); Supplementary Information PDF (MOESM1, 25 pages) | source checked ["source-hash-verification","pdf-text-parse","independent-cell-check","rendered-source-visual-check"] · 2026-10-11 author reported Audit detailsExtraction: Parsed the pdftotext -layout text of the pinned Supplementary Information PDF with tools/extract_ntbench.py: each block's 'Dataset' header gives six task columns, model rows matched as label followed by six 'value (± sd)' cells; block, row and column counts asserted. Supplementary Table 9, row 'NT 500M Randomly Initialized', approach 'Training from scratch', column 'Average performance' Independent review 2026-10-11: value, locator, metric, unit, direction and linked configuration and protocol checked; corrections from the review applied by the extractor as listed in review.md. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record Extraction artifact SHA-256: |
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
1 source record and release history
- Nucleotide Transformer: Supplementary Information (Supplementary Tables 1-12) · Original source · Nature Methods 22, 287-297 (published online 2024-11-28); Supplementary Information PDF (MOESM1, 25 pages)
Technical metadata and extraction receipts
Stable ID: lit-genomics-20261011-ntbench-result-nt-500m-randomly-initialized-training-from-scratch-summary
- metric
- matthews-correlation-coefficient
- metric direction
- higher
- unit
- unitless
- printed value
- 0.355
- source locator
- Supplementary Table 9, row 'NT 500M Randomly Initialized', approach 'Training from scratch', column 'Average performance'
- review
- method: source-hash-verification; pdf-text-parse; independent-cell-check; rendered-source-visual-check; method note: Matched the re-downloaded Supplementary Information PDF's SHA-256; an independent parser over the text layer compared all values (printed, numeric and SD) with 0 mismatches; a seeded stratified sample of 62 cells was checked against rendered pages 7 and 8.; reviewer: claude; reviewer note: Claude review agent reviewer-gen-2, independent of the extractor lit-genomics; no human review claimed; date: 2026-10-11; artifact sha256: ef0f357a1289aad1910059d455579e242a9a82ab3e32ec38d8ea2584effc436e; retrieval url: https://static-content.springer.com/esm/art%3A10.1038%2Fs41592-024-02523-z/MediaObjects/41592_2024_2523_MOESM1_ESM.pdf; note: Extraction: Parsed the pdftotext -layout text of the pinned Supplementary Information PDF with tools/extract_ntbench.py: each block's 'Dataset' header gives six task columns, model rows matched as label followed by six 'value (± sd)' cells; block, row and column counts asserted. Supplementary Table 9, row 'NT 500M Randomly Initialized', approach 'Training from scratch', column 'Average performance' Independent review 2026-10-11: value, locator, metric, unit, direction and linked configuration and protocol checked; corrections from the review applied by the extractor as listed in review.md.
- numeric value
- 0.355
- metric qualifier
- mean over three task categories of median cross-validation MCC
- missing metadata
- uncertainty: reason: unreported