作者: Ling Xue , Jeffrey W. Godden , Jürgen Bajorath
DOI: 10.1021/CI990308D
关键词:
摘要: In an effort to identify biologically active molecules in compound databases, we have investigated similarity searching using short binary bit strings with a maximum of 54 positions. These “minifingerprints” (MFPs) were designed account for the presence or absence structural fragments and/or aromatic character, flexibility, and hydrogen-bonding capacity molecules. MFP design was based on analysis distributions molecular descriptors two large collections. The performance different MFPs reference fingerprint tested by systematic “one-against-all” searches database containing 364 compounds biological activities. For each fingerprint, most effective cutoff value determined. An accounting only 32 showed less than 2% false positive matches correctly assigned average ∼40% same activity query ...