Dbtop: topomer similarity searching of conventional structure databases.

摘要: Abstract A new topomer-based method for 3D searching of conventional structural databases is described, according to which molecular structures are compared as sets fragments or topomers, in single rule-generated conformations oriented by superposition their fragmentation bonds. topomer characterized its CoMFA-like steric shape and now also pharmacophoric features, some novel ways that detailed discussed. To illustrate the behavior similarity searching, a dbtop program was used generate distance matrix diverse set 26 PDE4 inhibitors 15 serotonin receptor modulators. With best three parameter settings tried, within 210 shortest distances (of 1460), 94.7% involved pairs compounds having same biological activity, nearest neighbor every compound shared activity. The standard metric, Tanimoto coefficients “2D fingerprints”, could achieve similar selectivity performance only 108 distances, neighbors had different Topomer allowed “lead-hopping” among 22 inhibitors, notably between rolipram cipamfylline, while fingerprints” Tanimotos recognized generally classes. In 370 searches authentic high-throughput screening (HTS) data sets, typical search rate about 200 per s.