rewirebio.iobenchmarks
Configuration

Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)

Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model): DNABERT-2 backbone with LoRA on Wqkv, gated-attention MIL pooling and MLP; high/medium/low-TF sequential fine-tuning (final model) on ichorCNA tumour-fraction strata of the training cohort.

4 evaluations · 24 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.

Model Capabilities & Hardware Requirements

Deployment facts recorded for Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model). Fields the record does not carry are marked not recorded.

ParametersNot recorded
Minimum VRAMNot recorded
Inference LatencyNot recorded
Context WindowNot recorded
AccessNot recorded
LicenseNot recorded

Evaluations and results

4 evaluations · 24 results. Different protocols are not a single leaderboard.

Sorted by AUROC (all cancers vs non-cancer) (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 configurationProtocol and datasetAUROC (all cancers vs non-cancer)Recall (sensitivity, all cancers, at the predefined cutoff (98% specificity in the training cohort))AUROC (low-TF (< 3%) cancers vs non-cancer)Recall (sensitivity, low-TF (< 3%) stage I/II cancers, at the predefined cutoff (98% specificity in the training cohort))Recall (sensitivity, low-TF (< 3%) stage III/IV cancers, at the predefined cutoff (98% specificity in the training cohort))Specificity (non-cancer controls, at the predefined cutoff (98% specificity in the training cohort))AUROC (high-TF (> 5%) cancers vs non-cancer)AUROC (medium-TF (3-5%) cancers vs non-cancer)Recall (sensitivity, high-TF (> 5%) cancers, at the predefined cutoff (98% specificity in the training cohort))Recall (sensitivity, low-TF (< 3%) cancers, at the predefined cutoff (98% specificity in the training cohort))Recall (sensitivity, medium-TF (3-5%) cancers, at the predefined cutoff (98% specificity in the training cohort))
Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)EvaluationFragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction, Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)0.93 (best in column)0.580.91 (best in column)0.3910.7080.97 (best in column)0.9840.9580.910.460.72
Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)EvaluationFragmentia-AI WGS paper: External DELFI cohort (FinaleDB), DELFI cohort from FinaleDB as used by the Fragmentia-AI WGS paper (538 samples)0.9290.78 (best in column)Not reportedNot reportedNot reported0.92Not reportedNot reportedNot reportedNot reportedNot reported
Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)EvaluationFragmentia-AI WGS paper: Training cohort (resubstitution), Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)0.9240.580.9050.3960.716 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)EvaluationFragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction, Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)0.9170.610.8840.397 (best in column)0.683Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
All 24 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: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: External DELFI cohort (FinaleDB)
Dataset: DELFI cohort from FinaleDB as used by the Fragmentia-AI WGS paper (538 samples)
0.929 AUROC (all cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.907 to 0.951. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the external delfi cohort (finaledb)

ctdnalit-20261011-fragmentiawgs2026-protocol-delfi

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P21, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: External DELFI cohort (FinaleDB)
Dataset: DELFI cohort from FinaleDB as used by the Fragmentia-AI WGS paper (538 samples)
0.78 Recall (sensitivity, all cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the external delfi cohort (finaledb)

ctdnalit-20261011-fragmentiawgs2026-protocol-delfi

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P21, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: External DELFI cohort (FinaleDB)
Dataset: DELFI cohort from FinaleDB as used by the Fragmentia-AI WGS paper (538 samples)
0.92 Specificity (non-cancer controls, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the external delfi cohort (finaledb)

ctdnalit-20261011-fragmentiawgs2026-protocol-delfi

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P21, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.984 AUROC (high-TF (> 5%) cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.979 to 0.989. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.91 AUROC (low-TF (< 3%) cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.902 to 0.917. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.958 AUROC (medium-TF (3-5%) cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.940 to 0.976. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.93 AUROC (all cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.924 to 0.936. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 2
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.58 Recall (sensitivity, all cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.91 Recall (sensitivity, high-TF (> 5%) cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 6
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.46 Recall (sensitivity, low-TF (< 3%) cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 6
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.391 Recall (sensitivity, low-TF (< 3%) stage I/II cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.367 to 0.415. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Test' / 'I/II', column 'Sensitivity (95% CI)'
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.708 Recall (sensitivity, low-TF (< 3%) stage III/IV cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.666 to 0.748. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Test' / 'III/IV', column 'Sensitivity (95% CI)'
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.72 Recall (sensitivity, medium-TF (3-5%) cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 6
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Independent test cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS independent test cohort (5,996 samples, 0.05x ULP-WGS)
0.97 Specificity (non-cancer controls, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the independent test cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-test

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P19, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Training cohort (resubstitution)
Dataset: Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)
0.905 AUROC (low-TF (< 3%) cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.894 to 0.916. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the training cohort (resubstitution)

ctdnalit-20261011-fragmentiawgs2026-protocol-train

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Training cohort (resubstitution)
Dataset: Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)
0.924 AUROC (all cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.916 to 0.932. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the training cohort (resubstitution)

ctdnalit-20261011-fragmentiawgs2026-protocol-train

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 1
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Training cohort (resubstitution)
Dataset: Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)
0.58 Recall (sensitivity, all cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the training cohort (resubstitution)

ctdnalit-20261011-fragmentiawgs2026-protocol-train

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Training cohort (resubstitution)
Dataset: Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)
0.396 Recall (sensitivity, low-TF (< 3%) stage I/II cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.368 to 0.423. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the training cohort (resubstitution)

ctdnalit-20261011-fragmentiawgs2026-protocol-train

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Train' / 'I/II', column 'Sensitivity (95% CI)'
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Training cohort (resubstitution)
Dataset: Fragmentia-AI WGS training cohort (3,695 samples, 0.1x ULP-WGS)
0.716 Recall (sensitivity, low-TF (< 3%) stage III/IV cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.667 to 0.762. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the training cohort (resubstitution)

ctdnalit-20261011-fragmentiawgs2026-protocol-train

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Train' / 'III/IV', column 'Sensitivity (95% CI)'
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)
0.884 AUROC (low-TF (< 3%) cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.868 to 0.901. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the internal validation cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-validation

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 5
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)
0.917 AUROC (all cancers vs non-cancer)
unitless · higher

Uncertainty: 95% CI 0.905 to 0.929. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the internal validation cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-validation

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 1
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)
0.61 Recall (sensitivity, all cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the internal validation cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-validation

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Results P14, sentence 3
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)
0.397 Recall (sensitivity, low-TF (< 3%) stage I/II cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.350 to 0.445. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the internal validation cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-validation

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Validation' / 'I/II', column 'Sensitivity (95% CI)'
Configuration: Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model)Protocol: Fragmentia-AI WGS paper: Internal validation cohort by ichorCNA tumour fraction
Dataset: Fragmentia-AI WGS internal validation cohort (2,043 samples, 0.1x ULP-WGS)
0.683 Recall (sensitivity, low-TF (< 3%) stage III/IV cancers, at the predefined cutoff (98% specificity in the training cohort))
fraction · higher

Uncertainty: 95% CI 0.594 to 0.763. Method not stated

Coverage: Not reported scored / Not reported eligible

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

Fragmentia-AI WGS high/medium/low-TF sequential fine-tuning (final model) on the internal validation cohort by ichorcna tumour fraction

ctdnalit-20261011-fragmentiawgs2026-protocol-validation

Aggregation: Not reported

Generalizable cancer detection from ultra-low-pass WGS via deep contextual modeling of cfDNA sequences; Fragmentia-AI WGS paper, Supplementary Material 1 (Tables S1-S4, Figures S1-S9) · Supplementary Table S4, row 'Validation' / 'III/IV', column 'Sensitivity (95% CI)'

Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-c8bea25e5abb
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-c8bea25e5abb · Record review: source checked

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

Technical metadata for this record is tracked under stable ID ctdnalit-20261011-fragmentiawgs2026-config-high-medium-low-tf.

Source locator: Results P8, P13-P14; Methods P47-P60

Related records

How to cite

Fragmentia-AI WGS, high/medium/low-TF sequential fine-tuning (final model). Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/ctdnalit-20261011-fragmentiawgs2026-config-high-medium-low-tf/

Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.

The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence

Suggest a correction