Table 3 Comparisons with the state-of-the-art methods on independent datasets
From: DeepHLAPred: a deep learning-based method for non-classical HLA binder prediction
Datasets | DeepHLAPred | HLAncPred | MHCflurry 2.0 | NetMHCpan 4.1 | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
SN | SP | ACC | MCC | SN | SP | ACC | MCC | SN | SP | ACC | MCC | SN | SP | ACC | MCC | |
HLA-E*01:01 | 0.7195 | 0.8780 | 0.7988 | 0.6052 | 0.5854 | 0.9268 | 0.7561 | 0.5449 | 0.2804 | 0.9634 | 0.6220 | 0.3339 | 0.4756 | 0.9390 | 0.7073 | 0.4679 |
HLA-E*01:03 | 0.9402 | 0.8507 | 0.8955 | 0.7942 | 0.9552 | 0.8358 | 0.8955 | 0.7967 | 0.1493 | 0.9552 | 0.5522 | 0.1765 | 0.3880 | 0.9851 | 0.6866 | 0.4651 |