Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.832 Accuracy (Youden threshold from calibration set)fraction · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'Accuracy' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.674 AUPRCunitless · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'auPRC' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.896 AUROCunitless · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'auROC' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.846 Balanced accuracy (Youden threshold from calibration set)fraction · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'Balanced Acc.' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.631 MCC (Youden threshold from calibration set)unitless · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'MCC' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.5 Precision (Youden threshold from calibration set)fraction · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'Precision' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.867 Recall (Youden threshold from calibration set)fraction · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'Sensitivity' Configuration: Evo2 (Youden) (Mathur and Sachidanandam 2026) Protocol: Mitochondrial variant pathogenicity classification (Mathur and Sachidanandam 2026 Table 2) Dataset: Mitochondrial variant test split, MITOMAP pathogenic and ClinGen/ClinVar benign (Mathur and Sachidanandam 2026) 0.825 Specificity (Youden threshold from calibration set)fraction · higher
Uncertainty: Not reported by the source
Coverage: Not reported scored / Not reported eligible
Independent external evaluation · Source checked Evidence concern: excluded from comparisons Comparator rows reproduce APOGEE 2 (Bianco et al. 2023) cross-validation results on a missense-only set (140 pathogenic, 1,734 neutral). They were not measured on this paper's test split, so Table 2 does not compare like with like. Separately, Table 2 sensitivity/specificity 0.867/0.825 (balanced accuracy 0.846) versus Figure 2B and text 90.2%/85.0% (87.6%); the source does not explain the difference.Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Section 3 Methods (comparison paragraph, reference 27); Table 2; Figure 2B legend; Section 4.3.1
Methods, coverage and source Evo2 (Youden) on mtDNA pathogenicity test split
lit-genomics-20261011-evo2mito-protocol-mtdna-pathogenicity
Aggregation: Not reported
Benchmarking DNA Foundation Models: Biological Blind Spots in Evo2 Variant-Effect Prediction · Table 2 (PDF page 10), row 'Evo2 (Youden)', column 'Specificity'