Table 3. Exact classification results on the curated variant data.
Acceptor Gain | Acceptor Loss | Donor Gain | Donor Loss | |
---|---|---|---|---|
MES (Sliding Window) | 0.00% | 1.16% | 2.33% | 2.25% |
SpliceAI | 87.50% | 77.10% | 79.07% | 78.93% |
CI-SpliceAI | 93.75% | 78.55% | 79.07% | 82.02% |
Comparison of the accuracy of predicting the exact location of the splice disruption on the 388 variants with known exact disruption coordinates. Except for donor gains, where CI-SpliceAI and the original SpliceAI have equal accuracy, CI-SpliceAI outperforms other models.