From: Mixture of experts models to exploit global sequence similarity on biomolecular sequence labeling
Data Sets | Number of Sequences | Number of + Instances | Number of - Instances |
---|---|---|---|
RNA-prot 30% | 180 | 5398 | 27837 |
RNA-prot 60% | 215 | 6689 | 32073 |
RNA-prot 90% | 246 | 7798 | 34675 |
DNA-prot 30% | 257 | 5326 | 53494 |
DNA-prot 60% | 289 | 5974 | 58031 |
DNA-prot 90% | 317 | 6551 | 60877 |