Skip to main content

Advertisement

Table 4 This table indicates which features are contained in which feature sets

From: Predicting a small molecule-kinase interaction map: A machine learning approach

  FS1 FS2 FS3 FS4(C5) FS4(SVM) FS5 FS6(C5) FS6(SVM) FS7(C5) FS7(SVM)
PT X X X X X X X X X X
MS X X X X X X X X X X
FTs X X X X X X X X X X
KNN X X X X X X X X X X
CF          X X
GF          X X
P          X X
STTK X X X X X X X X X X
PC X X X X X X X X X X
PRO X X X X X X X X X X
Apri X X X X X X X X X X
glAli   X   X    X   X  
locAli    X X    X X X X
PSF      X X X X X X
abPSF       X X X X X
  1. This table indicates which features are contained in which feature sets. PT: Primary Targets, MS: 2D Molecular Structure, FTs: Free Trees, KNN: KNN clustering, CF: Chemical Features, GF: Geometric Features, P: Pharmacophores, STTK: Partitioning in Serine-, Threonine and Tyrosine Kinases, PC: Phylogenetic Clustering, PRO: PROSITE patterns, Apri: APriori patterns, glAli: global alignment scores, locAli: local alignment scores, PSF: Position Specific Features, abPSF: abstract Position Specific Features. The upper part of the table describes the chemical features, the lower part the biological features. In the left part of the table (from FS1 to FS6), the description of the kinases is optimized (testing combinations of alignment-based and position-specific features). In the right part (FS7), the chemical representation is further optimized by additional descriptors.