Multiple target screening method for robust and accurate in silico ligand screening

A Virtual Active Compound Produced from the Negative Image of a Ligand-binding Pocket, and its Application to in-silico Drug Screening

“…These methods have been applied to many known protein-ligand complexes, and correctly predict 40-70% of complex conformations within the 2 Å root-mean-square deviation (RMSD) [4,11]. In contrast, the prediction of protein-compound binding free energy is still poor; the average error of the prediction of free energy is almost 2.5 kcal/mol, which is close to the binding free energy of hit compounds [5,[12][13][14][15]. Recent docking programs work very fast; usually the program can predict the protein-compound affinity within one minute.…”

“…Several methods have been proposed to improve database enrichment, some of which are based on the protein-compound affinity matrix [11,[15][16][17]. One of these methods is the multiple active site correction (MASC) score [16].…”

“…The results obtained by the MASC scoring method and the RS method strongly depend on the computational parameters such as grid size, atom model, scoring system, etc. The multiple target screening (MTS) method, on the other hand, could provide stable screening results [15]. For the k-th compound, the docking scores {s i k ; i = 1,...,M} are sorted in descending order; here the suffix i represents the i-th pocket and the total number of pockets is M. The order n i k is assigned to each i-th pocket depending on its value s i k .…”

Finding ligands for G protein-coupled receptors based on the protein–compound affinity matrix

“…This approach is successful when the target protein structure is properly prepared and the size of the ligand is small [13][14][15][16][17][18][19][20]. The hit ratio of the conventional singletarget in silico screening is low, and several multiple-target screening methods have been proposed to improve it [12,21,22]. These approaches require the 3D atomic structure of the target protein, and are successful, based on the rapid increase in the number of entries in the Protein Data Bank (PDB).…”

Prediction of protein–ligand complex structure by docking software guided by other complex structures

“…Another method of improving database enrichment is the www.elsevier.com/locate/JMGM Journal of Molecular Graphics and Modelling 26 (2008) 1030-1033 application of the protein-compound affinity matrix. The multiple active site correction (MASC) scoring method uses the deviation of the docking score instead of the raw docking score [10], and the multiple target screening (MTS) method compares the docking scores of many proteins for one compound instead of comparing the docking scores of many compounds for one target protein [11,12]. These methods indirectly improve the estimation of the binding free energy, but do not improve docking accuracy.…”