Fast rule-based bioactivity prediction using associative classification mining

Table 3 Property descriptors used in the modeling

ADMET	ADMET_BBB_Level,ADMET_Absorption_Level,ADMET_CYP2D6,ADMET_PPB_Level
Physiochemical	ALogP,Molecular_Solubility,Molecular_SurfaceArea,Molecular_PolarSurfaceArea,Molecular_FractionalPolarSurfaceArea,Molecular_SASA,Molecular_PolarSASA,Molecular_FractionalPolarSASA,Molecular_SAVol,ChemAxon_LogP,ChemAxon_Polarizability,ChemAxon_Refractivity,ChemAxon_TPSA,FormalCharge
Simple counts	Num_Atoms,Num_Bonds,Num_Hydrogens,Num_NegativeAtoms,Num_RingBonds,Num_RotatableBonds,Num_BridgeBonds,Num_Rings,Num_RingAssemblies,Num_Chains,Num_ChainAssemblies,Molecular_Weight,Num_H_Acceptors,Num_H_Donors,ChemAxon_HBA,ChemAxon_HBD

Note: All property descriptors are computed by using Pipeline Pilot. The name and meaning of property descriptors can be found in Pipeline Pilot help documents. In most cases, the meaning of a name can be determined from the name itself. For example, ADMET_BBB_LEVEL means ranking of the LogBB values by using Accelrys blood–brain barrier penetration model: 0 is very high; 1 is high; 2 is medium; 3 is low and 4 is undefined, namely, molecule is outside of the confidence area of the regression model used to calculate LogBB.

ISSN: 1758-2946