rewirebio.iobenchmarks
Protocol

Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features

Five classifiers on the same 20,499 DMR-informed CpG features and the same three train/validation/test partitions; mean over split-runs.

5 evaluations · 15 results

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.

5 recorded evaluations, 15 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

5 evaluations · 15 results. Different protocols are not a single leaderboard.

Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features · Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)

Sorted by AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs) (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 configurationAUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
CpGPT Cancer-100M fine-tuned on 20,499 DMR-informed CpGsEvaluation0.996 (best in column)0.99 (best in column)0.949 (best in column)
SVM radial on 20,499 DMR-informed CpGs (caret)Evaluation0.9830.9570.937
Elastic Net on 20,499 DMR-informed CpGs (caret)Evaluation0.9810.9420.93
GBM on 20,499 DMR-informed CpGs (caret)Evaluation0.9550.8460.877
Random Forest on 20,499 DMR-informed CpGs (caret)Evaluation0.9480.8410.793
All 15 result rows with coverage, uncertainty and sources
Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: CpGPT Cancer-100M fine-tuned on 20,499 DMR-informed CpGsProtocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.996 AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.9950 to 0.9970. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Independent external evaluation · Source checked
Methods, coverage and source

CpGPT on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'CpGPT', column 'Test AUPRC'
Configuration: CpGPT Cancer-100M fine-tuned on 20,499 DMR-informed CpGsProtocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.99 AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.9715 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Independent external evaluation · Source checked
Methods, coverage and source

CpGPT on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'CpGPT', column 'Test AUROC'
Configuration: CpGPT Cancer-100M fine-tuned on 20,499 DMR-informed CpGsProtocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.949 AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.8887 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Independent external evaluation · Source checked
Methods, coverage and source

CpGPT on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'CpGPT', column 'Validation AUROC'
Configuration: Elastic Net on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.981 AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.9494 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Elastic Net on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Elastic Net', column 'Test AUPRC'
Configuration: Elastic Net on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.942 AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.8587 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Elastic Net on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Elastic Net', column 'Test AUROC'
Configuration: Elastic Net on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.93 AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.7928 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Elastic Net on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Elastic Net', column 'Validation AUROC'
Configuration: GBM on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.955 AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.9122 to 0.9978. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

GBM on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'GBM', column 'Test AUPRC'
Configuration: GBM on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.846 AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.7322 to 0.9602. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

GBM on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'GBM', column 'Test AUROC'
Configuration: GBM on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.877 AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.7917 to 0.9617. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

GBM on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'GBM', column 'Validation AUROC'
Configuration: Random Forest on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.948 AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.8985 to 0.9965. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Random Forest on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Random Forest', column 'Test AUPRC'
Configuration: Random Forest on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.841 AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.7573 to 0.9243. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Random Forest on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Random Forest', column 'Test AUROC'
Configuration: Random Forest on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.793 AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.5946 to 0.9920. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Random Forest on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'Random Forest', column 'Validation AUROC'
Configuration: SVM radial on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.983 AUPRC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.9488 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

SVM radial on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'SVM radial', column 'Test AUPRC'
Configuration: SVM radial on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.957 AUROC (GI cancer vs non-cancer, held-out test partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.8736 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

SVM radial on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'SVM radial', column 'Test AUROC'
Configuration: SVM radial on 20,499 DMR-informed CpGs (caret)Protocol: Xu and Feng 2026: GI cancer vs non-cancer on GSE149438, three train-validation-test split-runs, DMR-informed CpG features
Dataset: EpiPanGI Dx plasma cfDNA targeted bisulfite methylation (GSE149438, 300 samples)
0.937 AUROC (GI cancer vs non-cancer, validation partition, mean over three split-runs)
unitless · higher

Uncertainty: 95% CI 0.8126 to 1.0000. Across three split-runs, bounded to the valid range 0 to 1 (Table 3 footnote)

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

SVM radial on GSE149438 DMR-informed split-runs (Xu and Feng 2026)

ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs

Aggregation: Not reported

An Evaluation of DMR-Informed Fine-Tuning of Tissue Array-Pretrained CpGPT for Gastrointestinal Cancer Classification Using cfDNA Targeted Methylation · Table 3, row 'SVM radial', column 'Validation AUROC'

Source checking is not independent reproduction. Release 2026-11-21-adc15377fab2.

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.

Author-reported evaluations
4
External evaluations
1

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

Null control

Proposed control: requires review

Training-set class prior where supervised fitting is permitted

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

Regularised classifier on simple permitted features, or protocol's conventional reference

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 (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)

Coverage is derived from release 2026-11-21-adc15377fab2. 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.

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 2026-11-21-adc15377fab2
Property and statementOriginal source and locationReview and provenance

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

Sources and history

Release 2026-11-21-adc15377fab2 · Record review: source checked

1 source record and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID ctdnalit-20261011-cpgpt2026-protocol-dmr-informed-split-runs.

Source locator: Table 3; Methods; Results

Related records

Suggest a correction