Fig. 2
From: iIL13Pred: improved prediction of IL-13 inducing peptides using popular machine learning classifiers
The boxplot representation of the top 10 features selected by the mRMR feature selection method indicates their discriminatory nature: The top 10 features selected by the mRMR feature selection method were plotted for IL-13 inducing and non-IL-13 inducing peptides. The features were found to be highly discriminatory as seen in the box plot of the top 10 features. Abbreviations: mRMR, minimum redundancy maximum relevance; BTC_S, Composition of Single bonds; CeTD_SS1, Composition of group 1 residues for secondary structure attribute; TPC_RQF, Composition of Arginine–Glutamine–Phenylalanine tripeptide; SER_I, Shannon entropy for residue Isoleucine; SER_L, Shannon entropy for residue Leucine; SER_T, Shannon entropy for residue Threonine; BTC_T, Composition of total bonds; CeTD_75_p_HB1, Number of group 1 residues for hydrophobicity present in 75% quartile; AAC_H, Amino acid composition of Histidine; SER_P, Shannon entropy for residue Proline