| Configuration: ABC_A=DNase, C=Average ENCODE Hi-C (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.561 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.505742840168912 to 0.609499053593117 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceABC_A=DNase, C=Average ENCODE Hi-C on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 20 (D20:F20); Dataset 'Combined K562 (training)', Predictor 'ABC_A=DNase, C=Average ENCODE Hi-C', metric 'Weighted AUPRC' |
|---|
| Configuration: ABC_A=DNase, C=Average ENCODE Hi-C (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.453 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.413472264469473 to 0.492071262218147 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceABC_A=DNase, C=Average ENCODE Hi-C on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 26 (D26:F26); Dataset 'Combined K562 (training)', Predictor 'ABC_A=DNase, C=Average ENCODE Hi-C', metric 'Weighted precision at threshold' |
|---|
| Configuration: ABC_A=DNase, C=Average ENCODE Hi-C (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.739 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.69600194650038 to 0.782655613574707 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceABC_A=DNase, C=Average ENCODE Hi-C on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 32 (D32:F32); Dataset 'Combined K562 (training)', Predictor 'ABC_A=DNase, C=Average ENCODE Hi-C', metric 'Weighted recall at threshold' |
|---|
| Configuration: Distance to TSS (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.435 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.381294301517761 to 0.487138690854213 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceDistance to TSS on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 24 (D24:F24); Dataset 'Combined K562 (training)', Predictor 'Distance to TSS', metric 'Weighted AUPRC' |
|---|
| Configuration: Distance to TSS (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.291 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.262423650910323 to 0.319101531577943 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceDistance to TSS on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 30 (D30:F30); Dataset 'Combined K562 (training)', Predictor 'Distance to TSS', metric 'Weighted precision at threshold' |
|---|
| Configuration: Distance to TSS (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.755 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.712739109282484 to 0.795269968200388 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceDistance to TSS on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 36 (D36:F36); Dataset 'Combined K562 (training)', Predictor 'Distance to TSS', metric 'Weighted recall at threshold' |
|---|
| Configuration: ENCODE-rE2G (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.666 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.614835111980482 to 0.711557617708767 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceENCODE-rE2G on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 21 (D21:F21); Dataset 'Combined K562 (training)', Predictor 'ENCODE-rE2G', metric 'Weighted AUPRC' |
|---|
| Configuration: ENCODE-rE2G (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.548 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.502957160758294 to 0.58983971918054 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceENCODE-rE2G on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 27 (D27:F27); Dataset 'Combined K562 (training)', Predictor 'ENCODE-rE2G', metric 'Weighted precision at threshold' |
|---|
| Configuration: ENCODE-rE2G (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.73 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.68634462 to 0.77396333 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceENCODE-rE2G on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 33 (D33:F33); Dataset 'Combined K562 (training)', Predictor 'ENCODE-rE2G', metric 'Weighted recall at threshold' |
|---|
| Configuration: Correlation_E-P DNase-seq signal (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.188 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.153700202257717 to 0.226562560550906 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCorrelation_E-P DNase-seq signal on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 25 (D25:F25); Dataset 'Combined K562 (training)', Predictor 'Correlation_E-P DNase-seq signal', metric 'Weighted AUPRC' |
|---|
| Configuration: Correlation_E-P DNase-seq signal (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.0846 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.0748657 to 0.09422633 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCorrelation_E-P DNase-seq signal on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 31 (D31:F31); Dataset 'Combined K562 (training)', Predictor 'Correlation_E-P DNase-seq signal', metric 'Weighted precision at threshold' |
|---|
| Configuration: Correlation_E-P DNase-seq signal (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.71 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.663954244027132 to 0.754126177061213 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceCorrelation_E-P DNase-seq signal on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 37 (D37:F37); Dataset 'Combined K562 (training)', Predictor 'Correlation_E-P DNase-seq signal', metric 'Weighted recall at threshold' |
|---|
| Configuration: EpiMap (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.311 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.261110315175289 to 0.361017256946356 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEpiMap on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 23 (D23:F23); Dataset 'Combined K562 (training)', Predictor 'EpiMap', metric 'Weighted AUPRC' |
|---|
| Configuration: EpiMap (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.187 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.166855358972052 to 0.2065905167216 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEpiMap on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 29 (D29:F29); Dataset 'Combined K562 (training)', Predictor 'EpiMap', metric 'Weighted precision at threshold' |
|---|
| Configuration: EpiMap (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.714 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.669049563420841 to 0.757391291403457 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEpiMap on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 35 (D35:F35); Dataset 'Combined K562 (training)', Predictor 'EpiMap', metric 'Weighted recall at threshold' |
|---|
| Configuration: EPIraction (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.446 AUPRC (weighted by estimated probability of a direct cis-effect) fraction · higher Uncertainty: 95% CI 0.39383868 to 0.500921087804116 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEPIraction on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 22 (D22:F22); Dataset 'Combined K562 (training)', Predictor 'EPIraction', metric 'Weighted AUPRC' |
|---|
| Configuration: EPIraction (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.381 Precision (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.345706679909842 to 0.414719708782195 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEPIraction on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 28 (D28:F28); Dataset 'Combined K562 (training)', Predictor 'EPIraction', metric 'Weighted precision at threshold' |
|---|
| Configuration: EPIraction (Gschwind et al. 2026) | Protocol: Combined K562 CRISPR training pairs, weighted metrics (Gschwind et al. 2026 Supplementary Table 3) Dataset: MPRabc K562 CRISPRi benchmark | 0.749 Recall (weighted; at predictor threshold) fraction · higher Uncertainty: 95% CI 0.705437423576133 to 0.791130313360929 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceEPIraction on combined K562 CRISPR training pairs regulatory-variant-20261009-protocol-gschwind2026-k562-training-weighted Aggregation: Not reported An encyclopedia of human enhancer-gene regulatory interactions; Gschwind et al. 2026, Supplementary Table 3 · Supplementary Table 3 'Held-out benchmarks', row 34 (D34:F34); Dataset 'Combined K562 (training)', Predictor 'EPIraction', metric 'Weighted recall at threshold' |
|---|
| Configuration: Activity-by-Contact (ABC) — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.612 (0.567–0.656) AUPRC fraction · higher Uncertainty: 95% CI 0.567 to 0.656 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc abc Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Results, Figure 2 comparison paragraph, Activity-by-Contact (ABC), auprc |
|---|
| Configuration: MPRabc — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.707 (0.663–0.746) AUPRC fraction · higher Uncertainty: 95% CI 0.663 to 0.746 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc full Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Table 3, MPRabc, auprc |
|---|
| Configuration: MPRabc — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.680 (0.637–0.724) Precision at 70% recall fraction · higher Uncertainty: 95% CI 0.637 to 0.724 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc full Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Table 3, MPRabc, precision-at-70-recall |
|---|
| Configuration: MPRabc megamap Hi-C — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.69 AUPRC fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc megamap Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Results, Figure 2 comparison paragraph, MPRabc megamap Hi-C, auprc |
|---|
| Configuration: MPRabc (MPRALegNet only; no Sei) — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.686 (0.642–0.728) AUPRC fraction · higher Uncertainty: 95% CI 0.642 to 0.728 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc mpralegnet Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Table 3, MPRabc (MPRALegNet only; no Sei), auprc |
|---|
| Configuration: MPRabc (MPRALegNet only; no Sei) — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.655 (0.610–0.695) Precision at 70% recall fraction · higher Uncertainty: 95% CI 0.610 to 0.695 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc mpralegnet Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Table 3, MPRabc (MPRALegNet only; no Sei), precision-at-70-recall |
|---|
| Configuration: ENCODE-rE2G — K562 published comparison | Protocol: MPRabc K562 CRISPRi enhancer–gene evaluation Dataset: MPRabc K562 CRISPRi benchmark | 0.634 (0.587–0.684) AUPRC fraction · higher Uncertainty: 95% CI 0.587 to 0.684 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcemprabc re2g Not reported Aggregation: Not reported MPRA-informed modeling improves context-specific enhancer–gene regulatory interactions · Table 3, ENCODE-rE2G, auprc |
|---|