Clustering and Rule-Based Classifications of Chemical Structures Evaluated in the Biological Activity Space

详细信息    查看全文

• 作者：Ansgar Schuffenhauer ; Nathan Brown ; Peter Ertl ; Jeremy L. Jenkins ; Paul Selzer ; Jacques Hamon
• 刊名：Journal of Chemical Information and Modeling
• 出版年：2007
• 出版时间：March 2007
• 年：2007
• 卷：47
• 期：2
• 页码：325 - 336
• 全文大小：503K
• 年卷期：v.47,no.2(March 2007)
• ISSN：1549-960X

文摘

Classification methods for data sets of molecules according to their chemical structure were evaluated fortheir biological relevance, including rule-based, scaffold-oriented classification methods and clustering basedon molecular descriptors. Three data sets resulting from uniformly determined in vitro biological profilingexperiments were classified according to their chemical structures, and the results were compared in a Paretoanalysis with the number of classes and their average spread in the profile space as two concurrent objectiveswhich were to be minimized. It has been found that no classification method is overall superior to all otherstudied methods, but there is a general trend that rule-based, scaffold-oriented methods are the better choiceif classes with homogeneous biological activity are required, but a large number of clusters can be tolerated.On the other hand, clustering based on chemical fingerprints is superior if fewer and larger classes arerequired, and some loss of homogeneity in biological activity can be accepted.