Figure 3.

The F-scores of features and relevant distribution. A. F-scores of the first 20 dimensional features that are the standard amino acid composition. The distribution of the 7th and 13th feature with star marked, F7 and F13 which are the composition of amino acid of Ile and Gln, is illustrated in D and E. B. F-scores of the next 21 dimensional features that are the hydrophobicity of protein sequence. The first three dimensions, F21-23, represent compositions of hydrophobic, neutral and polar amino acid, respectively; F24-26, represent the percentage of transition frequency between hydrophobic, neutral and polar amino acid residues; F27-41 are which the first, 25%, 50%, 75% and 100% of the amino acids of a particular group are located. The distribution of feature F21 and F24 with high F-score are shown in F and G. C. The F-scores of complete 146 features.