Database mining with adaptive fuzzy partition: Application to the prediction of pesticide toxicity on rats

Marco Pintorie, Nadège Piclin, Emilio Benfenati, Giuseppina Gini, Jacquies R. Chrétien

Research output: Contribution to journalArticle

Abstract

A data set of 235 pesticide compounds, divided into three classes according to their toxicity toward rats, was analyzed by a fuzzy logic procedure called adaptive fuzzy partition (AFP). This method allows the establishment of molecular descriptor/chemical activity relationships by dynamically dividing the descriptor space into a set of fuzzily partitioned subspaces. A set of 153 molecular descriptors was analyzed, including topological, physicochemical, quantum mechanical, constitutional, and electronic parameters, and the most relevant descriptors were selected with the help of a procedure combining genetic algorithm concepts and a stepwise method. The ability of this AFP model to classify the three toxicity classes was validated after dividing the data set compounds into training and test sets, including 165 and 70 molecules, respectively. The experimental class was correctly predicted for 76% of the test-set compounds. Furthermore, the most toxic class, particularly important for real applications of the toxicity models, was correctly predicted in 86% of cases. Finally, a comparison between the results obtained by AFP and those obtained by other classic classification techniques showed that AFP improved the predictive power of the proposed models.

Original languageEnglish
Pages (from-to)983-991
Number of pages9
JournalEnvironmental Toxicology and Chemistry
Volume22
Issue number5
DOIs
Publication statusPublished - May 1 2003

Fingerprint

Pesticides
Toxicity
Rats
Databases
toxicity
prediction
Poisons
fuzzy mathematics
Genetic Phenomena
Fuzzy Logic
genetic algorithm
Fuzzy logic
pesticide
Genetic algorithms
Molecules
toxicity of pesticides
method
test
Datasets

Keywords

  • Fuzzy logic
  • Genetic algorithms
  • Pesticide toxicity
  • Rats
  • Structure-activity relationships

ASJC Scopus subject areas

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

Cite this

Database mining with adaptive fuzzy partition : Application to the prediction of pesticide toxicity on rats. / Pintorie, Marco; Piclin, Nadège; Benfenati, Emilio; Gini, Giuseppina; Chrétien, Jacquies R.

In: Environmental Toxicology and Chemistry, Vol. 22, No. 5, 01.05.2003, p. 983-991.

Research output: Contribution to journalArticle

Pintorie, Marco ; Piclin, Nadège ; Benfenati, Emilio ; Gini, Giuseppina ; Chrétien, Jacquies R. / Database mining with adaptive fuzzy partition : Application to the prediction of pesticide toxicity on rats. In: Environmental Toxicology and Chemistry. 2003 ; Vol. 22, No. 5. pp. 983-991.
@article{5d729dc6cc734b7ebccf63bf0d600384,
title = "Database mining with adaptive fuzzy partition: Application to the prediction of pesticide toxicity on rats",
abstract = "A data set of 235 pesticide compounds, divided into three classes according to their toxicity toward rats, was analyzed by a fuzzy logic procedure called adaptive fuzzy partition (AFP). This method allows the establishment of molecular descriptor/chemical activity relationships by dynamically dividing the descriptor space into a set of fuzzily partitioned subspaces. A set of 153 molecular descriptors was analyzed, including topological, physicochemical, quantum mechanical, constitutional, and electronic parameters, and the most relevant descriptors were selected with the help of a procedure combining genetic algorithm concepts and a stepwise method. The ability of this AFP model to classify the three toxicity classes was validated after dividing the data set compounds into training and test sets, including 165 and 70 molecules, respectively. The experimental class was correctly predicted for 76{\%} of the test-set compounds. Furthermore, the most toxic class, particularly important for real applications of the toxicity models, was correctly predicted in 86{\%} of cases. Finally, a comparison between the results obtained by AFP and those obtained by other classic classification techniques showed that AFP improved the predictive power of the proposed models.",
keywords = "Fuzzy logic, Genetic algorithms, Pesticide toxicity, Rats, Structure-activity relationships",
author = "Marco Pintorie and Nad{\`e}ge Piclin and Emilio Benfenati and Giuseppina Gini and Chr{\'e}tien, {Jacquies R.}",
year = "2003",
month = "5",
day = "1",
doi = "10.1897/1551-5028(2003)022<0983:DMWAFP>2.0.CO;2",
language = "English",
volume = "22",
pages = "983--991",
journal = "Environmental Toxicology and Chemistry",
issn = "0730-7268",
publisher = "John Wiley and Sons Ltd",
number = "5",

}

TY - JOUR

T1 - Database mining with adaptive fuzzy partition

T2 - Application to the prediction of pesticide toxicity on rats

AU - Pintorie, Marco

AU - Piclin, Nadège

AU - Benfenati, Emilio

AU - Gini, Giuseppina

AU - Chrétien, Jacquies R.

PY - 2003/5/1

Y1 - 2003/5/1

N2 - A data set of 235 pesticide compounds, divided into three classes according to their toxicity toward rats, was analyzed by a fuzzy logic procedure called adaptive fuzzy partition (AFP). This method allows the establishment of molecular descriptor/chemical activity relationships by dynamically dividing the descriptor space into a set of fuzzily partitioned subspaces. A set of 153 molecular descriptors was analyzed, including topological, physicochemical, quantum mechanical, constitutional, and electronic parameters, and the most relevant descriptors were selected with the help of a procedure combining genetic algorithm concepts and a stepwise method. The ability of this AFP model to classify the three toxicity classes was validated after dividing the data set compounds into training and test sets, including 165 and 70 molecules, respectively. The experimental class was correctly predicted for 76% of the test-set compounds. Furthermore, the most toxic class, particularly important for real applications of the toxicity models, was correctly predicted in 86% of cases. Finally, a comparison between the results obtained by AFP and those obtained by other classic classification techniques showed that AFP improved the predictive power of the proposed models.

AB - A data set of 235 pesticide compounds, divided into three classes according to their toxicity toward rats, was analyzed by a fuzzy logic procedure called adaptive fuzzy partition (AFP). This method allows the establishment of molecular descriptor/chemical activity relationships by dynamically dividing the descriptor space into a set of fuzzily partitioned subspaces. A set of 153 molecular descriptors was analyzed, including topological, physicochemical, quantum mechanical, constitutional, and electronic parameters, and the most relevant descriptors were selected with the help of a procedure combining genetic algorithm concepts and a stepwise method. The ability of this AFP model to classify the three toxicity classes was validated after dividing the data set compounds into training and test sets, including 165 and 70 molecules, respectively. The experimental class was correctly predicted for 76% of the test-set compounds. Furthermore, the most toxic class, particularly important for real applications of the toxicity models, was correctly predicted in 86% of cases. Finally, a comparison between the results obtained by AFP and those obtained by other classic classification techniques showed that AFP improved the predictive power of the proposed models.

KW - Fuzzy logic

KW - Genetic algorithms

KW - Pesticide toxicity

KW - Rats

KW - Structure-activity relationships

UR - http://www.scopus.com/inward/record.url?scp=0242416628&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242416628&partnerID=8YFLogxK

U2 - 10.1897/1551-5028(2003)022<0983:DMWAFP>2.0.CO;2

DO - 10.1897/1551-5028(2003)022<0983:DMWAFP>2.0.CO;2

M3 - Article

C2 - 12729207

AN - SCOPUS:0242416628

VL - 22

SP - 983

EP - 991

JO - Environmental Toxicology and Chemistry

JF - Environmental Toxicology and Chemistry

SN - 0730-7268

IS - 5

ER -