Resumen

CASTANETA MARONI, Heriberto et al. Análisis de clustering subspace de pesticidas químicos. Rev. Bol. Quim [online]. 2024, vol.41, n.3, pp.21-38.  Epub 12-Dic-2024. ISSN 0250-5460.  https://doi.org/10.34098/2078-3949.41.3.4.

Subspace clustering analysis of chemical pesticides. The “subspace clustering” algorithm was applied to analyse molecules with pesticidal properties. 1509 molecules from the PPDB (Pesticides Properties DataBase - AERU Hertfordshire University) database were analysed; 1005 molecules presented experimental vapour pressure data. Descriptors were calculated with the PaDEL-Descriptor program (v. 2.20) with a total of 14464 0D-2D molecular descriptors and fingerprint types. Subspace clustering allowed us to group molecules into clusters and simultaneously detect the descriptor subspaces that characterize each cluster. This technique allowed us to analyse the structure of a dataset by examining the similarity between groups of objects described in different subspaces, demonstrating its ability to study high-dimensional data.

Palabras clave : Subspace clustering; Molecular Descriptors; Vapor pressure; Pesticides; PPDB database; Quantitative Structure-Property Relationships; QSPR.

        · resumen en Español     · texto en Español     · Español ( pdf )