Advanced search
Start date
Betweenand
Related content
(Reference retrieved automatically from Web of Science through information on FAPESP grant and its corresponding number as mentioned in the publication by the authors.)

Nonequivalent Effects of Diverse LogP Algorithms in Three QSAR Studies

Full text
Author(s):
de Melo, Eduardo Borges [1, 2] ; Castro Ferreira, Marcia Miguel [1]
Total Authors: 2
Affiliation:
[1] Univ Estadual Campinas, Inst Chem, Theoret & Appl Chemometr Lab, BR-13083970 Campinas, SP - Brazil
[2] Univ Estadual Oeste Parana, Ctr Ciencias Med & Farmaceut, Curso Farm, BR-85819110 Cascavel, Parana - Brazil
Total Affiliations: 2
Document type: Journal article
Source: QSAR & COMBINATORIAL SCIENCE; v. 28, n. 10, p. 1156-1165, OCT 2009.
Web of Science Citations: 6
Abstract

Despite of the availability and facility of accessing several algorithms for calculation of LogP in QSA(P)R studies, articles typically do not describe the selection procedure for the method used. Therefore, three studies to verify the influence of different LogP algorithms on building QSAR models were performed. Two QSAR data sets from the literature (forty-two tricyclic phtalimide inhibitors of HIV-integrase and fourty-six TIBO derivatives inhibitors of HIV-reverse transcriptase) were used together with LogP calculated by thirteen algorithms, and several regression models were constructed and compared. A new QSAR study for 4,5-dihydroxvpyrimidine carboxamides inhibitors of HIV-1 integrase was also performed. The explained and predicted variance, results from external validation, leave-N-out cross-validation and y-randomization test were analyzed for all models from the three data sets. Despite the same physicochemical meaning, LogP's calculated by distinct methods may show different levels of contribution to the model. This observation comes out from the comparison of validated models. These results indicate that the arbitrary choice of one specific algorithm for LogP calculation, as is usual in QSA(P)R studies, does not necessarily lead to the highest quality model for the analyzed data set. (AU)

FAPESP's process: 04/04686-5 - Classic and new chemiometric approaches in theoretic studies of bioactive substances
Grantee:Marcia Miguel Castro Ferreira
Support Opportunities: Research Projects - Thematic Grants