QSAR modelling of water quality indices of alkylphenol pollutants

J. H. Kim, P. Gramatica, M. G. Kim, D. Kim, Paul Tratnyek

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The aim of this study was to determine the degradability of 26 Alkylphenols (APs) by Chemical Oxygen Demand (COD) and/or 5-day Biochemical Oxygen Demand (BOD5), and to describe these data from Quantitative Structure-activity Relationships (QSARs). Statistical analysis techniques, such as Multiple Linear Regression (MLR), Principal Component Regression (PCR), Partial Least-Squares (PLS) Regression and Neural Network (NN) were carried out to calibrate and validate four-descriptor QSAR models using two different types of descriptor sets. Stable MLR-QSAR models using Leave-One-Out (LOO) were obtained with high predictability performance: r2 = 0.924, Q cv 2 = 0.854 for log (1/BOD) model on 24 APs and r 2 = 0.888, Qcv 2 = 0.818 for log (1/COD) on all the studied APs. The MLR models, built with four Dragon descriptors selected by Genetic Algorithm (GA), presented the following performances on 24 APs: r 2 = 0.889, Qcv 2 = 0.848 for log (1/BOD5) and r2 = 0.885, Qcv 2 = 0.834 for log (1/COD) on 26 compounds. From these results, it is expected that the QSAR models generated could be successfully expanded to predict the biological and chemical activities of structurally diverse AP compounds.

Original languageEnglish (US)
Pages (from-to)729-743
Number of pages15
JournalSAR and QSAR in Environmental Research
Volume18
Issue number7-8
DOIs
StatePublished - Dec 2007

Fingerprint

water quality
Quantitative Structure-Activity Relationship
Water Quality
Biological Oxygen Demand Analysis
Water quality
contaminants
regression analysis
Linear Models
Chemical oxygen demand
biochemical oxygen demand
Linear regression
pollutant
chemical oxygen demand
Biochemical oxygen demand
modeling
oxygen
Least-Squares Analysis
activity (biology)
genetic algorithms
genetic algorithm

Keywords

  • Alkylphenols
  • BOD5
  • COD
  • MLR
  • QSAR

ASJC Scopus subject areas

  • Chemistry(all)
  • Physical and Theoretical Chemistry
  • Environmental Science(all)
  • Toxicology
  • Health, Toxicology and Mutagenesis

Cite this

QSAR modelling of water quality indices of alkylphenol pollutants. / Kim, J. H.; Gramatica, P.; Kim, M. G.; Kim, D.; Tratnyek, Paul.

In: SAR and QSAR in Environmental Research, Vol. 18, No. 7-8, 12.2007, p. 729-743.

Research output: Contribution to journalArticle

Kim, J. H. ; Gramatica, P. ; Kim, M. G. ; Kim, D. ; Tratnyek, Paul. / QSAR modelling of water quality indices of alkylphenol pollutants. In: SAR and QSAR in Environmental Research. 2007 ; Vol. 18, No. 7-8. pp. 729-743.
@article{855f15a7be154daeb7b67c736773dbc4,
title = "QSAR modelling of water quality indices of alkylphenol pollutants",
abstract = "The aim of this study was to determine the degradability of 26 Alkylphenols (APs) by Chemical Oxygen Demand (COD) and/or 5-day Biochemical Oxygen Demand (BOD5), and to describe these data from Quantitative Structure-activity Relationships (QSARs). Statistical analysis techniques, such as Multiple Linear Regression (MLR), Principal Component Regression (PCR), Partial Least-Squares (PLS) Regression and Neural Network (NN) were carried out to calibrate and validate four-descriptor QSAR models using two different types of descriptor sets. Stable MLR-QSAR models using Leave-One-Out (LOO) were obtained with high predictability performance: r2 = 0.924, Q cv 2 = 0.854 for log (1/BOD) model on 24 APs and r 2 = 0.888, Qcv 2 = 0.818 for log (1/COD) on all the studied APs. The MLR models, built with four Dragon descriptors selected by Genetic Algorithm (GA), presented the following performances on 24 APs: r 2 = 0.889, Qcv 2 = 0.848 for log (1/BOD5) and r2 = 0.885, Qcv 2 = 0.834 for log (1/COD) on 26 compounds. From these results, it is expected that the QSAR models generated could be successfully expanded to predict the biological and chemical activities of structurally diverse AP compounds.",
keywords = "Alkylphenols, BOD5, COD, MLR, QSAR",
author = "Kim, {J. H.} and P. Gramatica and Kim, {M. G.} and D. Kim and Paul Tratnyek",
year = "2007",
month = "12",
doi = "10.1080/10629360701698761",
language = "English (US)",
volume = "18",
pages = "729--743",
journal = "SAR and QSAR in Environmental Research",
issn = "1062-936X",
publisher = "Taylor and Francis Ltd.",
number = "7-8",

}

TY - JOUR

T1 - QSAR modelling of water quality indices of alkylphenol pollutants

AU - Kim, J. H.

AU - Gramatica, P.

AU - Kim, M. G.

AU - Kim, D.

AU - Tratnyek, Paul

PY - 2007/12

Y1 - 2007/12

N2 - The aim of this study was to determine the degradability of 26 Alkylphenols (APs) by Chemical Oxygen Demand (COD) and/or 5-day Biochemical Oxygen Demand (BOD5), and to describe these data from Quantitative Structure-activity Relationships (QSARs). Statistical analysis techniques, such as Multiple Linear Regression (MLR), Principal Component Regression (PCR), Partial Least-Squares (PLS) Regression and Neural Network (NN) were carried out to calibrate and validate four-descriptor QSAR models using two different types of descriptor sets. Stable MLR-QSAR models using Leave-One-Out (LOO) were obtained with high predictability performance: r2 = 0.924, Q cv 2 = 0.854 for log (1/BOD) model on 24 APs and r 2 = 0.888, Qcv 2 = 0.818 for log (1/COD) on all the studied APs. The MLR models, built with four Dragon descriptors selected by Genetic Algorithm (GA), presented the following performances on 24 APs: r 2 = 0.889, Qcv 2 = 0.848 for log (1/BOD5) and r2 = 0.885, Qcv 2 = 0.834 for log (1/COD) on 26 compounds. From these results, it is expected that the QSAR models generated could be successfully expanded to predict the biological and chemical activities of structurally diverse AP compounds.

AB - The aim of this study was to determine the degradability of 26 Alkylphenols (APs) by Chemical Oxygen Demand (COD) and/or 5-day Biochemical Oxygen Demand (BOD5), and to describe these data from Quantitative Structure-activity Relationships (QSARs). Statistical analysis techniques, such as Multiple Linear Regression (MLR), Principal Component Regression (PCR), Partial Least-Squares (PLS) Regression and Neural Network (NN) were carried out to calibrate and validate four-descriptor QSAR models using two different types of descriptor sets. Stable MLR-QSAR models using Leave-One-Out (LOO) were obtained with high predictability performance: r2 = 0.924, Q cv 2 = 0.854 for log (1/BOD) model on 24 APs and r 2 = 0.888, Qcv 2 = 0.818 for log (1/COD) on all the studied APs. The MLR models, built with four Dragon descriptors selected by Genetic Algorithm (GA), presented the following performances on 24 APs: r 2 = 0.889, Qcv 2 = 0.848 for log (1/BOD5) and r2 = 0.885, Qcv 2 = 0.834 for log (1/COD) on 26 compounds. From these results, it is expected that the QSAR models generated could be successfully expanded to predict the biological and chemical activities of structurally diverse AP compounds.

KW - Alkylphenols

KW - BOD5

KW - COD

KW - MLR

KW - QSAR

UR - http://www.scopus.com/inward/record.url?scp=36448938278&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36448938278&partnerID=8YFLogxK

U2 - 10.1080/10629360701698761

DO - 10.1080/10629360701698761

M3 - Article

C2 - 18038370

AN - SCOPUS:36448938278

VL - 18

SP - 729

EP - 743

JO - SAR and QSAR in Environmental Research

JF - SAR and QSAR in Environmental Research

SN - 1062-936X

IS - 7-8

ER -