Premium
Pattern recognition and neural networks applied to structure–activity relationships of neolignans tested against Leishmania amazonensis using quantum chemical and topological descriptors
Author(s) -
Costa Maria Cristina Andreazza,
Barata Lauro Euclides Soares,
Bergmann Bartira Rossi,
Takahata Yuji
Publication year - 2003
Publication title -
international journal of quantum chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.484
H-Index - 105
eISSN - 1097-461X
pISSN - 0020-7608
DOI - 10.1002/qua.10698
Subject(s) - quantum chemical , artificial neural network , topology (electrical circuits) , pattern recognition (psychology) , quantum , molecule , chemistry , leishmania , set (abstract data type) , biological system , computational chemistry , artificial intelligence , mathematics , physics , computer science , biology , combinatorics , quantum mechanics , organic chemistry , parasite hosting , world wide web , programming language
Pattern recognition (PR) and neural networks (NN) were applied to structure–activity relationship studies of a series of neolignans tested against Leishmania amazonensis. Comparison of NN with PR methods revealed that the capability of the two methods are similar to classify the molecules in the different categories, but it was found that K‐nearest neighbors and NN are superior to SIMCA in the activity prevision of a new set of compounds. The applicability of these methods using quantum chemical descriptors and topological indices were investigated. © 2003 Wiley Periodicals, Inc. Int J Quantum Chem 95: 295–302, 2003