On the use of PLS and N-PLS in MIA-QSAR: Azole antifungals

Mohammad Goodarzi; Matheus P. Freitas

doi:10.1016/j.chemolab.2008.11.007

On the use of PLS and N-PLS in MIA-QSAR: Azole antifungals

Mohammad Goodarzi, Matheus P. Freitas

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

30 Scopus citations

Abstract

The antifungal activities of a series of azole derivatives have been modeled by using MIA (multivariate image analysis) descriptors. Two regression methods were applied to correlate such descriptors with the activities column vector: bilinear (classical) and multilinear (N-way) partial least squares - PLS and N-PLS, respectively. The PLS-based model for this series of compounds demonstrated higher predictive ability than the N-PLS-based model, in opposition to some published results for other series of compounds. The activities block was taken in logarithmic scale (pMIC_90(cpd)/pMIC_{90(bifonazole)}) and the statistical performance of both models was found to be significantly better than the CoMFA analysis previously established.

Original language	English (US)
Pages (from-to)	59-62
Number of pages	4
Journal	Chemometrics and Intelligent Laboratory Systems
Volume	96
Issue number	1
DOIs	https://doi.org/10.1016/j.chemolab.2008.11.007
State	Published - Mar 15 2009

Keywords

Antifungals
MIA-QSAR
N-PLS regression
PLS regression

ASJC Scopus subject areas

Analytical Chemistry
Software
Process Chemistry and Technology
Spectroscopy
Computer Science Applications

Access to Document

10.1016/j.chemolab.2008.11.007

Cite this

@article{080b3b6f0abb4af2b6e86a076655e101,

title = "On the use of PLS and N-PLS in MIA-QSAR: Azole antifungals",

abstract = "The antifungal activities of a series of azole derivatives have been modeled by using MIA (multivariate image analysis) descriptors. Two regression methods were applied to correlate such descriptors with the activities column vector: bilinear (classical) and multilinear (N-way) partial least squares - PLS and N-PLS, respectively. The PLS-based model for this series of compounds demonstrated higher predictive ability than the N-PLS-based model, in opposition to some published results for other series of compounds. The activities block was taken in logarithmic scale (pMIC90(cpd)/pMIC90(bifonazole)) and the statistical performance of both models was found to be significantly better than the CoMFA analysis previously established.",

keywords = "Antifungals, MIA-QSAR, N-PLS regression, PLS regression",

author = "Mohammad Goodarzi and Freitas, {Matheus P.}",

note = "Funding Information: FAPEMIG (grant number: CEX 415/06) is gratefully acknowledged for the support of this research, as is CNPq for the fellowship (M.P.F.).",

year = "2009",

month = mar,

day = "15",

doi = "10.1016/j.chemolab.2008.11.007",

language = "English (US)",

volume = "96",

pages = "59--62",

journal = "Chemometrics and Intelligent Laboratory Systems",

issn = "0169-7439",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - On the use of PLS and N-PLS in MIA-QSAR

T2 - Azole antifungals

AU - Goodarzi, Mohammad

AU - Freitas, Matheus P.

N1 - Funding Information: FAPEMIG (grant number: CEX 415/06) is gratefully acknowledged for the support of this research, as is CNPq for the fellowship (M.P.F.).

PY - 2009/3/15

Y1 - 2009/3/15

N2 - The antifungal activities of a series of azole derivatives have been modeled by using MIA (multivariate image analysis) descriptors. Two regression methods were applied to correlate such descriptors with the activities column vector: bilinear (classical) and multilinear (N-way) partial least squares - PLS and N-PLS, respectively. The PLS-based model for this series of compounds demonstrated higher predictive ability than the N-PLS-based model, in opposition to some published results for other series of compounds. The activities block was taken in logarithmic scale (pMIC90(cpd)/pMIC90(bifonazole)) and the statistical performance of both models was found to be significantly better than the CoMFA analysis previously established.

AB - The antifungal activities of a series of azole derivatives have been modeled by using MIA (multivariate image analysis) descriptors. Two regression methods were applied to correlate such descriptors with the activities column vector: bilinear (classical) and multilinear (N-way) partial least squares - PLS and N-PLS, respectively. The PLS-based model for this series of compounds demonstrated higher predictive ability than the N-PLS-based model, in opposition to some published results for other series of compounds. The activities block was taken in logarithmic scale (pMIC90(cpd)/pMIC90(bifonazole)) and the statistical performance of both models was found to be significantly better than the CoMFA analysis previously established.

KW - Antifungals

KW - MIA-QSAR

KW - N-PLS regression

KW - PLS regression

UR - http://www.scopus.com/inward/record.url?scp=61349178836&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=61349178836&partnerID=8YFLogxK

U2 - 10.1016/j.chemolab.2008.11.007

DO - 10.1016/j.chemolab.2008.11.007

M3 - Article

AN - SCOPUS:61349178836

SN - 0169-7439

VL - 96

SP - 59

EP - 62

JO - Chemometrics and Intelligent Laboratory Systems

JF - Chemometrics and Intelligent Laboratory Systems

IS - 1

ER -

On the use of PLS and N-PLS in MIA-QSAR: Azole antifungals

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this