A Fragment-weighted Key-based Similarity Measure for Use in Structural Clustering and Virtual Screening

Autor:	Klaus Gundertofte, Jan Torleif Pedersen, Morten Langgård, Anne Marie Jørgensen
Rok vydání:	2006
Předmět:	Virtual screening Fuzzy clustering Computer science Organic Chemistry Single-linkage clustering Correlation clustering Similarity measure computer.software_genre Computer Science Applications Set (abstract data type) Similarity (network science) Drug Discovery Data mining Cluster analysis computer
Zdroj:	QSAR & Combinatorial Science. 25:221-234
ISSN:	1611-0218 1611-020X
DOI:	10.1002/qsar.200510157
Popis:	A new similarity measure and structural clustering method has been developed in which each structure is fragmented into ring systems, linkers, and side chains, and where the similarity between structures forms a weighted sum of the similarities between each set of fragments. We have applied the MACCS keys as molecular descriptors and identified a number of ways in which to improve the use of these keys for structural clustering and virtual screening. The method has been optimized to reproduce scaffold-biased clustering commonly used in medicinal chemistry.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::4dcf3a7fd7bea4f4bd408e22f664ce26 https://doi.org/10.1002/qsar.200510157 Zobrazit plný text záznamu Plný text