PLS and N-PLS-based MIA-QSTR modelling of the acute toxicities of phenylsulphonyl carboxylates to Vibrio fischeri

Mohammad Goodarzi, Matheus P. Freitas

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Descriptors based on multivariate image analysis have been used to derive predictive QSTR models of the acute toxicities of phenylsulphonylcarboxylates to Vibrio fischeri. Classical and multilinear partial least squares, PLS and N-PLS, respectively, were applied as regression methods, demonstrating similar predictive capability to each other. Model performance was improved in c. 10% after removing an outlier, and the results were in general agreement with the ones previously obtained through CoMFA and extended topochemical atom indices analysis. Overall, this study showed that a simple procedure is able to give highly predictive models, useful in ecotoxicology, independent of the regression method used for this class of compounds.

Original languageEnglish (US)
Pages (from-to)953-959
Number of pages7
JournalMolecular Simulation
Volume36
Issue number12
DOIs
StatePublished - Oct 1 2010

Fingerprint

Predictive Model
Toxicity
toxicity
Acute
carboxylates
CoMFA
Multivariate Image Analysis
Regression
regression analysis
Partial Least Squares
Performance Model
Modeling
Descriptors
Outlier
image analysis
Image analysis
Atoms
atoms
Class

Keywords

  • MIA-QSTR
  • N-PLS
  • phenylsulphonyl carboxylates
  • PLS
  • Vibrio fischeri

ASJC Scopus subject areas

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

Cite this

PLS and N-PLS-based MIA-QSTR modelling of the acute toxicities of phenylsulphonyl carboxylates to Vibrio fischeri. / Goodarzi, Mohammad; Freitas, Matheus P.

In: Molecular Simulation, Vol. 36, No. 12, 01.10.2010, p. 953-959.

Research output: Contribution to journalArticle

@article{ca8f95ad6fe842298d211f265c409d4a,
title = "PLS and N-PLS-based MIA-QSTR modelling of the acute toxicities of phenylsulphonyl carboxylates to Vibrio fischeri",
abstract = "Descriptors based on multivariate image analysis have been used to derive predictive QSTR models of the acute toxicities of phenylsulphonylcarboxylates to Vibrio fischeri. Classical and multilinear partial least squares, PLS and N-PLS, respectively, were applied as regression methods, demonstrating similar predictive capability to each other. Model performance was improved in c. 10{\%} after removing an outlier, and the results were in general agreement with the ones previously obtained through CoMFA and extended topochemical atom indices analysis. Overall, this study showed that a simple procedure is able to give highly predictive models, useful in ecotoxicology, independent of the regression method used for this class of compounds.",
keywords = "MIA-QSTR, N-PLS, phenylsulphonyl carboxylates, PLS, Vibrio fischeri",
author = "Mohammad Goodarzi and Freitas, {Matheus P.}",
year = "2010",
month = "10",
day = "1",
doi = "10.1080/08927022.2010.492836",
language = "English (US)",
volume = "36",
pages = "953--959",
journal = "Molecular Simulation",
issn = "0892-7022",
publisher = "Taylor and Francis Ltd.",
number = "12",

}

TY - JOUR

T1 - PLS and N-PLS-based MIA-QSTR modelling of the acute toxicities of phenylsulphonyl carboxylates to Vibrio fischeri

AU - Goodarzi, Mohammad

AU - Freitas, Matheus P.

PY - 2010/10/1

Y1 - 2010/10/1

N2 - Descriptors based on multivariate image analysis have been used to derive predictive QSTR models of the acute toxicities of phenylsulphonylcarboxylates to Vibrio fischeri. Classical and multilinear partial least squares, PLS and N-PLS, respectively, were applied as regression methods, demonstrating similar predictive capability to each other. Model performance was improved in c. 10% after removing an outlier, and the results were in general agreement with the ones previously obtained through CoMFA and extended topochemical atom indices analysis. Overall, this study showed that a simple procedure is able to give highly predictive models, useful in ecotoxicology, independent of the regression method used for this class of compounds.

AB - Descriptors based on multivariate image analysis have been used to derive predictive QSTR models of the acute toxicities of phenylsulphonylcarboxylates to Vibrio fischeri. Classical and multilinear partial least squares, PLS and N-PLS, respectively, were applied as regression methods, demonstrating similar predictive capability to each other. Model performance was improved in c. 10% after removing an outlier, and the results were in general agreement with the ones previously obtained through CoMFA and extended topochemical atom indices analysis. Overall, this study showed that a simple procedure is able to give highly predictive models, useful in ecotoxicology, independent of the regression method used for this class of compounds.

KW - MIA-QSTR

KW - N-PLS

KW - phenylsulphonyl carboxylates

KW - PLS

KW - Vibrio fischeri

UR - http://www.scopus.com/inward/record.url?scp=78249279374&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78249279374&partnerID=8YFLogxK

U2 - 10.1080/08927022.2010.492836

DO - 10.1080/08927022.2010.492836

M3 - Article

AN - SCOPUS:78249279374

VL - 36

SP - 953

EP - 959

JO - Molecular Simulation

JF - Molecular Simulation

SN - 0892-7022

IS - 12

ER -