| Configuration: Zeng et al. 2026 classifier: Combined All Features (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.97 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0227978278651072. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined All Features (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 2 'Combined All Features (1%)', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined All Features (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.86 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.0981509647480165 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined All Features (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 2 'Combined All Features (1%)', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined All Features (5%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.932 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0424929753478305. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined All Features (5%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 3 'Combined All Features (5%)', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined All Features (5%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.789 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.107405582907328 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined All Features (5%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 3 'Combined All Features (5%)', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: End Motifs | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.93 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0519161667413912. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEnd Motifs on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 7 'End Motifs', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: End Motifs | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.826 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.136261133071318 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEnd Motifs on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 7 'End Motifs', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Fragment Ratio | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.923 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0705965005812822. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceFragment Ratio on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 8 'Fragment Ratio', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Fragment Ratio | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.795 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.193024274189439 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceFragment Ratio on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 8 'Fragment Ratio', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Insert Size | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.748 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0974579710205907. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceInsert Size on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 9 'Insert Size', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Insert Size | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.283 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.19494005154084 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceInsert Size on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 9 'Insert Size', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Methylation | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.986 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0209780603284754. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceMethylation on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 10 'Methylation', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Methylation | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.949 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.0870136746761039 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceMethylation on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 10 'Methylation', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Motif + Methylation (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.938 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0401311333432108. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Motif + Methylation (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 4 'Combined Motif + Methylation (1%)', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Motif + Methylation (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.773 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.14899523978531 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Motif + Methylation (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 4 'Combined Motif + Methylation (1%)', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Motif + Methylation + Ratios (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.974 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0242400530902103. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Motif + Methylation + Ratios (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 5 'Combined Motif + Methylation + Ratios (1%)', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Motif + Methylation + Ratios (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.896 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.0857671794470052 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Motif + Methylation + Ratios (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 5 'Combined Motif + Methylation + Ratios (1%)', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Nucleosome Peak | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.689 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.1141088789247. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNucleosome Peak on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 11 'Nucleosome Peak', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Nucleosome Peak | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.397 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.143496781034909 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNucleosome Peak on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 11 'Nucleosome Peak', column B 'mean_sens99' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Ratios + Methylation (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.961 AUROC (cancer vs healthy, primary PE dataset, mean over 10 fold models) unitless · higher Uncertainty: SD 0.0300613681834156. Figure 4 legend: error bars in a, c, e and g are ±1 s.d. across cross-validation folds (n = 10). Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Ratios + Methylation (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4a, row 6 'Combined Ratios + Methylation (1%)', column B 'mean_auc' |
|---|
| Configuration: Zeng et al. 2026 classifier: Combined Ratios + Methylation (1%) | Protocol: Zeng et al. 2026 primary PE dataset: cancer vs healthy, nested cross-validation Dataset: Zeng et al. 2026 primary paired-end cfMeDIP-seq dataset (420 samples) | 0.857 Sensitivity at 99% specificity (cancer vs healthy, primary PE dataset, mean over 10 fold models) fraction · higher Uncertainty: ± 0.0828537049838524 (type not stated). Source Data columns are headed sd_sens99 and the sheets' text column prints 'Sensitivity=x ± y'; the Figure 4 legend says the error bars in panels b, d, f and h are s.e. Methods P61 says error bars denote across-fold s.d. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCombined Ratios + Methylation (1%) on pe primary cv (Zeng et al. 2026) ctdnalit-20261011-zeng2026-protocol-pe-primary-cv Aggregation: Not reported A pan-cancer compendium of 1,294 plasma cell-free DNA methylomes and fragmentomes enabling multicancer detection; Zeng et al. 2026, Source Data Fig. 4 (classifier AUC and sensitivity summaries) · Source Data Fig. 4, sheet Fig.4b, row 6 'Combined Ratios + Methylation (1%)', column B 'mean_sens99' |
|---|