Skip to main content

Table 2 Performance comparison among five strategies on four single subtype datasets

From: A context-free encoding scheme of protein sequences for predicting antigenicity of diverse influenza A viruses

Dataset Methods Accuracy Precision Recall F-score
H1N1 Liao et al. 0.742 0.717 0.877 0.788
  MutCounts 0.824 0.802 0.884 0.840
  Peng et al. 0.661 0.671 0.711 0.683
  RegionBand 0.706 0.669 red0.901 0.766
  CFreeEnS ared0.859 red0.856 0.887 red0.870
H3N2 Liao et al. 0.784 0.748 0.891 0.812
  MutCounts 0.843 0.841 0.851 0.845
  Peng et al. 0.720 0.658 red0.950 0.777
  RegionBand 0.790 0.763 0.864 0.809
  CFreeEnS red0.885 red0.896 0.882 red0.889
H5N1 Liao et al. 0.753 0.758 0.878 0.813
  MutCounts 0.863 0.859 0.915 0.885
  Peng et al. 0.846 0.857 0.908 0.880
  RegionBand 0.858 0.824 red0.978 0.893
  CFreeEnS red0.915 red0.903 0.965 red0.932
H9N2 Liao et al. 0.708 0.816 0.819 0.810
  MutCounts 0.775 0.823 0.914 0.859
  Peng et al. 0.633 red0.888 0.601 0.702
  RegionBand 0.804 0.818 0.954 0.880
  CFreeEnS red0.850 0.860 red0.964 red0.908
  1. aThe highest scores among five strategies on each dataset are colored redred