Dataset

Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity

Held-out protein clusters from the cDNA display proteolysis set as curated by Dieckhaus et al.

Research readiness

These checks assess whether the evidence supports a reproducible investigation. A source-checked score alone does not meet these requirements.

Release 2026-10-10-6e93f504adfc · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing
annotations: verification is missing
dependence: verification is missing

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
recipe pinned: verification is missing
resource estimate: verification is missing

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
independent validation: verification is missing
overlap checked: verification is missing

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Read reviewed discrepancy investigations

Evaluation results

12 evaluations · 36 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: ACDC-NN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.52 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN', column 'Megascale PCC'
Configuration: ACDC-NN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	1.05 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN', column 'Megascale RMSE (kcal/mol)'
Configuration: ACDC-NN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.45 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN', column 'Megascale SCC'
Configuration: ACDC-NN-Seq (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.48 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN-Seq on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN-Seq', column 'Megascale PCC'
Configuration: ACDC-NN-Seq (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	1.08 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN-Seq on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN-Seq', column 'Megascale RMSE (kcal/mol)'
Configuration: ACDC-NN-Seq (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.44 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ACDC-NN-Seq on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ACDC-NN-Seq', column 'Megascale SCC'
Configuration: FoldX (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.4 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source FoldX on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'FoldX', column 'Megascale PCC'
Configuration: FoldX (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	3.87 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source FoldX on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'FoldX', column 'Megascale RMSE (kcal/mol)'
Configuration: FoldX (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.57 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source FoldX on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'FoldX', column 'Megascale SCC'
Configuration: MAESTRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.55 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MAESTRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MAESTRO', column 'Megascale PCC'
Configuration: MAESTRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	1.04 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MAESTRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MAESTRO', column 'Megascale RMSE (kcal/mol)'
Configuration: MAESTRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.47 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MAESTRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MAESTRO', column 'Megascale SCC'
Configuration: mCSM (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.49 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source mCSM on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'mCSM', column 'Megascale PCC'
Configuration: mCSM (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.97 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source mCSM on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'mCSM', column 'Megascale RMSE (kcal/mol)'
Configuration: mCSM (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.41 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source mCSM on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'mCSM', column 'Megascale SCC'
Configuration: MUPRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.31 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MUPRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MUPRO', column 'Megascale PCC'
Configuration: MUPRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	1.08 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MUPRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MUPRO', column 'Megascale RMSE (kcal/mol)'
Configuration: MUPRO (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.29 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MUPRO on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'MUPRO', column 'Megascale SCC'
Configuration: PROSTATA (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.64 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source PROSTATA on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'PROSTATA †', column 'Megascale PCC'
Configuration: PROSTATA (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.83 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source PROSTATA on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'PROSTATA †', column 'Megascale RMSE (kcal/mol)'
Configuration: PROSTATA (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.59 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source PROSTATA on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'PROSTATA †', column 'Megascale SCC'
Configuration: ProteinMPNN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.43 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ProteinMPNN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ProteinMPNN', column 'Megascale PCC'
Configuration: ProteinMPNN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	1.3 root-mean-squared-error kilocalorie-per-mole · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ProteinMPNN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ProteinMPNN', column 'Megascale RMSE (kcal/mol)'
Configuration: ProteinMPNN (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.49 spearman-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ProteinMPNN on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'ProteinMPNN', column 'Megascale SCC'
Configuration: RaSP (Dieckhaus et al. 2024)	Protocol: Megascale held-out test split ddG prediction (Dieckhaus et al. 2024 Table 2) Dataset: Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity	0.71 pearson-correlation unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RaSP on Megascale held-out test split (Dieckhaus et al. 2024) protein-stability-20261009-protocol-dieckhaus2024-megascale Aggregation: Not reported Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Table 2, row 'RaSP †', column 'Megascale PCC'

Source checking is not independent reproduction. Release 2026-10-10-6e93f504adfc.

Dataset and evaluation context

A dataset supplies biological observations. The evaluation protocol defines how those observations are split, used and scored.

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.

6 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-10-6e93f504adfc
Property and statement	Original source and location	Review and provenance
attributes.denominator 28312 Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.denominator` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.population 28,312 single-point mutations in test-split proteins; reliable ddG only; natural and de novo designed domains of 40-72 residues Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.population` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.source_locator Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.source_locator` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.split MMseqs2 clusters at 25% identity; no test protein has a homologue in the Megascale or Fireprot training sets Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `attributes.split` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
description Held-out protein clusters from the cDNA display proteolysis set as curated by Dieckhaus et al. Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `description` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
name Megascale (Tsuboyama 2023) test split, homology-clustered at 25% identity Context-only references	Transfer learning to leverage larger datasets for improved prediction of protein stability changes Original source ↗ Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods' Version: PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML Retrieved: 2026-10-09T21:01:06Z	not individually reviewed No individual claim review recorded Audit details Field: `name` Source artifact SHA-256: `ba9a763c388eeea47d70fd0d8e2fbf497f61fc8a88dc93d5dac261572daa8010` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

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

Sources and history

Release 2026-10-10-6e93f504adfc · Record review: source checked

1 source records and release history

Transfer learning to leverage larger datasets for improved prediction of protein stability changes · Original source · PNAS 121(6):e2314853121, published 2024-01-29; PMC10861915 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: protein-stability-20261009-data-dieckhaus2024-megascale-test

areas: proteins-complexes
contexts: research
population: 28,312 single-point mutations in test-split proteins; reliable ddG only; natural and de novo designed domains of 40-72 residues
split: MMseqs2 clusters at 25% identity; no test protein has a homologue in the Megascale or Fireprot training sets
denominator: 28312
source locator: Methods 'Datasets' paragraphs 2 and 4; Results 'Comparison with literature methods'
missing metadata: version: reason: unreported; note: Tsuboyama et al. 2023 release used; exact file version not stated

Related records

Suggest a correction