List of features used for predicting peptide detectability

Reference: H. Tang, R. J. Arnold, P. Alves, Z. Xun, D. E. Clemmer, M. V. Novotny, J. P. Reilly and P. Radivojac, A computational approach toward label-free protein quantification using predicted peptide detectability. ISMB (Supplement of Bioinformatics) 2006: 481-488.


Some features were pre-calculated using 14 methods as following:

  • 4 protein disorder predictors (ref: PMID 12910457) server implementation): 1) VL2; 2) VL2V; 3) VL2C; 4) VL2S.
  • 5) VLXT protein disorder predictor (ref: PMID: 11093259)
  • 6) Flexibility (ref: PMID: 8090708)
  • Hydrophobic moments with different parameters: 7) window=11, angle=100deg; 8) window=11, angle=160deg; 9) window=11, angle=120deg; (ref: PMID 6582470).
  • 10) VL3 disorder predictor (ref: PMID: 14579347 and PMID: 15751111, and server implementation)
  • 11) B-factor predictor (ref: PMID: 14691223)
  • 12) DisPhos predictor (ref: PMID: 14960716 and server implementation)
  • Calmodulin binding site predictors: 13) residue level; 14) region level; (ref: PMID: 16493654).

    • All 175 features include:

  • 42 features: 3 features for each method 1-14 calculated in +/-5, +/-10 and +/-15 windows;
  • 3 features for the maximal value calculated by method 11 in +/-5, +/-10 and +/-15 windows;
  • 20 features for the amino acid content of the peptide;
  • 20 features for the first amino acid residue of the peptide;
  • 20 features for the last amino acid residue of the peptide;
  • 20 features for the existance of an amino acid residue in the first three residues of the peptide;
  • 20 features for the existance of an amino acid residue in the last three residues of the peptide;
  • 20 features for the existance of an amino acid residue in the protein right after the peptide;
  • 2 features: if the peptide is at N- or C- terminal of the protein;
  • 2 features: if the peptide is within 50 residues from the N- or C- terminal of the protein;
  • 1 feature: the frequency of the hydrophobic amino acid residues;
  • 1 feature: the frequency of the aromatic amino acid residues;
  • 1 feature: entropy of the peptide
  • 1 feature: mass / length ratio;
  • 1 feature: mass of the peptide;
  • 1 feature: length of the peptide;