Automated search for potentially active compounds by using cluster trees

A new agglomerative (bottom-up) hierarchical cluster technique referred to as the Adaptive Mean-Linkage algorithm is derived. Cluster algorithms are also offered as a tool to explore the descriptor space knowing the quantitative structure–activity relationship (QSAR). The substituents are clustered...

Full description

Saved in:
Bibliographic Details
Published in:European journal of medicinal chemistry 1999-02, Vol.34 (2), p.83-92
Main Authors: Santos Magalhães, Nereide Stela, De Holanda Cavalcanti, Socrates Cabral, Alencar De Menezes, Irwin Rose, De Sousa Araújo, Adriano Antunes, Magalhães De Oliveira, Hélio, Alves, Antonio José
Format: Article
Language:English
Subjects:
adaptive mean-linkage algorithm
Antineoplastic agents
Biological and medical sciences
cluster analysis
General aspects
Medical sciences
Pharmacology. Drug treatments
phenylguanidines
QSAR
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A new agglomerative (bottom-up) hierarchical cluster technique referred to as the Adaptive Mean-Linkage algorithm is derived. Cluster algorithms are also offered as a tool to explore the descriptor space knowing the quantitative structure–activity relationship (QSAR). The substituents are clustered building a dendrogram (cluster tree) per site. Choosing appropriate pathways on such cluster trees according to the QSAR equation, an automated search for potentially active substituted compounds can be performed. Applications to a series of substituted phenylguanidines with anticancer activity are focused illustrating this approach.
ISSN:0223-5234
1768-3254
DOI:10.1016/S0223-5234(99)80043-9