MIA-QSAR modelling of activities of a series of AZT analogues: Bi- and multilinear PLS regression

Mohammad Goodarzi; Matheus Puggina De Freitas

doi:10.1080/08927020903278001

MIA-QSAR modelling of activities of a series of AZT analogues: Bi- and multilinear PLS regression

Mohammad Goodarzi, Matheus Puggina De Freitas

Research output: Contribution to journal › Article › peer-review

14 Scopus citations

Abstract

The activities of a series of azidothymidine derivatives, compounds with anti-HIV potency, were computationally modelled using multivariate image analysis applied to quantitative structure-activity relationships (MIA-QSAR). Two regression methods were tested in order to find the best correlation between actual and predicted activities: bilinear (traditional) partial least squares (PLS), applied to the unfolded dataset, and multilinear PLS (N-PLS), applied to the three-way array. The predictive abilities of the PLS- and N-PLS-based models were found to be nearly equivalent, and both the methods derived QSAR models that are statistically superior to conventional QSAR, in which physicochemical descriptors and multiple linear regression were applied.

Original language	English (US)
Pages (from-to)	267-272
Number of pages	6
Journal	Molecular Simulation
Volume	36
Issue number	4
DOIs	https://doi.org/10.1080/08927020903278001
State	Published - Apr 2010

Keywords

AZT analogues
HIV
MIA-QSAR
N-PLS
PLS

ASJC Scopus subject areas

Chemistry(all)
Information Systems
Modeling and Simulation
Chemical Engineering(all)
Materials Science(all)
Condensed Matter Physics

Access to Document

10.1080/08927020903278001

Cite this

@article{8e801d48379149f5826f06fe93533927,

title = "MIA-QSAR modelling of activities of a series of AZT analogues: Bi- and multilinear PLS regression",

abstract = "The activities of a series of azidothymidine derivatives, compounds with anti-HIV potency, were computationally modelled using multivariate image analysis applied to quantitative structure-activity relationships (MIA-QSAR). Two regression methods were tested in order to find the best correlation between actual and predicted activities: bilinear (traditional) partial least squares (PLS), applied to the unfolded dataset, and multilinear PLS (N-PLS), applied to the three-way array. The predictive abilities of the PLS- and N-PLS-based models were found to be nearly equivalent, and both the methods derived QSAR models that are statistically superior to conventional QSAR, in which physicochemical descriptors and multiple linear regression were applied.",

keywords = "AZT analogues, HIV, MIA-QSAR, N-PLS, PLS",

author = "Mohammad Goodarzi and {De Freitas}, {Matheus Puggina}",

note = "Funding Information: CNPq is gratefully acknowledged for the fellowship (to M.P.F.), and FAPEMIG of Brazil and the Consejo Nacional de Investigaciones Cient{\'i}ficas y T{\'e}cnicas (CONICET) of Argentina for the financial support.",

year = "2010",

month = apr,

doi = "10.1080/08927020903278001",

language = "English (US)",

volume = "36",

pages = "267--272",

journal = "Molecular Simulation",

issn = "0892-7022",

publisher = "Taylor and Francis Ltd.",

number = "4",

}

TY - JOUR

T1 - MIA-QSAR modelling of activities of a series of AZT analogues

T2 - Bi- and multilinear PLS regression

AU - Goodarzi, Mohammad

AU - De Freitas, Matheus Puggina

N1 - Funding Information: CNPq is gratefully acknowledged for the fellowship (to M.P.F.), and FAPEMIG of Brazil and the Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) of Argentina for the financial support.

PY - 2010/4

Y1 - 2010/4

N2 - The activities of a series of azidothymidine derivatives, compounds with anti-HIV potency, were computationally modelled using multivariate image analysis applied to quantitative structure-activity relationships (MIA-QSAR). Two regression methods were tested in order to find the best correlation between actual and predicted activities: bilinear (traditional) partial least squares (PLS), applied to the unfolded dataset, and multilinear PLS (N-PLS), applied to the three-way array. The predictive abilities of the PLS- and N-PLS-based models were found to be nearly equivalent, and both the methods derived QSAR models that are statistically superior to conventional QSAR, in which physicochemical descriptors and multiple linear regression were applied.

AB - The activities of a series of azidothymidine derivatives, compounds with anti-HIV potency, were computationally modelled using multivariate image analysis applied to quantitative structure-activity relationships (MIA-QSAR). Two regression methods were tested in order to find the best correlation between actual and predicted activities: bilinear (traditional) partial least squares (PLS), applied to the unfolded dataset, and multilinear PLS (N-PLS), applied to the three-way array. The predictive abilities of the PLS- and N-PLS-based models were found to be nearly equivalent, and both the methods derived QSAR models that are statistically superior to conventional QSAR, in which physicochemical descriptors and multiple linear regression were applied.

KW - AZT analogues

KW - HIV

KW - MIA-QSAR

KW - N-PLS

KW - PLS

UR - http://www.scopus.com/inward/record.url?scp=77949480630&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949480630&partnerID=8YFLogxK

U2 - 10.1080/08927020903278001

DO - 10.1080/08927020903278001

M3 - Article

AN - SCOPUS:77949480630

SN - 0892-7022

VL - 36

SP - 267

EP - 272

JO - Molecular Simulation

JF - Molecular Simulation

IS - 4

ER -

MIA-QSAR modelling of activities of a series of AZT analogues: Bi- and multilinear PLS regression

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this