Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition

Marco Pintore, Nadège Piclin, Emilio Benfenati, Giuseppina Gini, Jacques R. Chrétien

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Recent progress in the development of powerful tools suitable to design and to classify large chemical libraries can be fruitfully extended also to the ecotoxicity domain. Amidst these methods, Fuzzy Logic concepts, based on the possibility to handle the "concept of partial truth", constitute interesting approaches to developing general predictive models. In this work, a global strategy of Database Mining was applied on a data set of 568 chemicals, extracted from a toxicity database concerning the fathead minnow and divided into four classes, according to the toxicity ranges defined by the European Community legislation. Two large sets of molecular descriptors were tested on the 2D and 3D structures, and the best ones were selected with help of a procedure combining Genetic Algorithm concepts and stepwise method. After selecting the training set with a rational selection based on the Self Organizing Maps (SOM), structural-activity models were built by Adaptive Fuzzy Partition (AFP). This method consists in modeling relations between molecular descriptors and biological activities, by dynamically dividing the molecular descriptor hyperspace into a set of fuzzy subspaces. The best model was selected by a validation set, and its robustness was confirmed by predicting a test set of 80 chemicals never used to define the AFP models. An encouraging validation ratio of about 72% was obtained in the prediction of the experimental toxicity class. Furthermore, very similar results were obtained by using molecular descriptors computed on 2D or 3D structures.

Original languageEnglish
Pages (from-to)210-219
Number of pages10
JournalQSAR and Combinatorial Science
Volume22
Issue number2
Publication statusPublished - Apr 2003

Fingerprint

Molecular Descriptors
Fuzzy Partition
Cyprinidae
Toxicity
Fuzzy Subspace
Genetic Phenomena
Databases
Fuzzy Logic
Small Molecule Libraries
Hyperspace
Structural Models
Self organizing maps
Predictive Model
European Union
Self-organizing Map
Test Set
Bioactivity
Legislation
Large Set
Fuzzy logic

Keywords

  • Ecotoxicity
  • Fathead minnow
  • Fuzzy logic
  • SAR models

ASJC Scopus subject areas

  • Discrete Mathematics and Combinatorics
  • Pharmacology

Cite this

Pintore, M., Piclin, N., Benfenati, E., Gini, G., & Chrétien, J. R. (2003). Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition. QSAR and Combinatorial Science, 22(2), 210-219.

Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition. / Pintore, Marco; Piclin, Nadège; Benfenati, Emilio; Gini, Giuseppina; Chrétien, Jacques R.

In: QSAR and Combinatorial Science, Vol. 22, No. 2, 04.2003, p. 210-219.

Research output: Contribution to journalArticle

Pintore, M, Piclin, N, Benfenati, E, Gini, G & Chrétien, JR 2003, 'Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition', QSAR and Combinatorial Science, vol. 22, no. 2, pp. 210-219.
Pintore, Marco ; Piclin, Nadège ; Benfenati, Emilio ; Gini, Giuseppina ; Chrétien, Jacques R. / Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition. In: QSAR and Combinatorial Science. 2003 ; Vol. 22, No. 2. pp. 210-219.
@article{9f9c1ebff14e41d99b4ac8bf63837f90,
title = "Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition",
abstract = "Recent progress in the development of powerful tools suitable to design and to classify large chemical libraries can be fruitfully extended also to the ecotoxicity domain. Amidst these methods, Fuzzy Logic concepts, based on the possibility to handle the {"}concept of partial truth{"}, constitute interesting approaches to developing general predictive models. In this work, a global strategy of Database Mining was applied on a data set of 568 chemicals, extracted from a toxicity database concerning the fathead minnow and divided into four classes, according to the toxicity ranges defined by the European Community legislation. Two large sets of molecular descriptors were tested on the 2D and 3D structures, and the best ones were selected with help of a procedure combining Genetic Algorithm concepts and stepwise method. After selecting the training set with a rational selection based on the Self Organizing Maps (SOM), structural-activity models were built by Adaptive Fuzzy Partition (AFP). This method consists in modeling relations between molecular descriptors and biological activities, by dynamically dividing the molecular descriptor hyperspace into a set of fuzzy subspaces. The best model was selected by a validation set, and its robustness was confirmed by predicting a test set of 80 chemicals never used to define the AFP models. An encouraging validation ratio of about 72{\%} was obtained in the prediction of the experimental toxicity class. Furthermore, very similar results were obtained by using molecular descriptors computed on 2D or 3D structures.",
keywords = "Ecotoxicity, Fathead minnow, Fuzzy logic, SAR models",
author = "Marco Pintore and Nad{\`e}ge Piclin and Emilio Benfenati and Giuseppina Gini and Chr{\'e}tien, {Jacques R.}",
year = "2003",
month = "4",
language = "English",
volume = "22",
pages = "210--219",
journal = "QSAR and Combinatorial Science",
issn = "1611-020X",
publisher = "Wiley - VCH Verlag GmbH & CO. KGaA",
number = "2",

}

TY - JOUR

T1 - Predicting toxicity against the fathead minnow by Adaptive Fuzzy Partition

AU - Pintore, Marco

AU - Piclin, Nadège

AU - Benfenati, Emilio

AU - Gini, Giuseppina

AU - Chrétien, Jacques R.

PY - 2003/4

Y1 - 2003/4

N2 - Recent progress in the development of powerful tools suitable to design and to classify large chemical libraries can be fruitfully extended also to the ecotoxicity domain. Amidst these methods, Fuzzy Logic concepts, based on the possibility to handle the "concept of partial truth", constitute interesting approaches to developing general predictive models. In this work, a global strategy of Database Mining was applied on a data set of 568 chemicals, extracted from a toxicity database concerning the fathead minnow and divided into four classes, according to the toxicity ranges defined by the European Community legislation. Two large sets of molecular descriptors were tested on the 2D and 3D structures, and the best ones were selected with help of a procedure combining Genetic Algorithm concepts and stepwise method. After selecting the training set with a rational selection based on the Self Organizing Maps (SOM), structural-activity models were built by Adaptive Fuzzy Partition (AFP). This method consists in modeling relations between molecular descriptors and biological activities, by dynamically dividing the molecular descriptor hyperspace into a set of fuzzy subspaces. The best model was selected by a validation set, and its robustness was confirmed by predicting a test set of 80 chemicals never used to define the AFP models. An encouraging validation ratio of about 72% was obtained in the prediction of the experimental toxicity class. Furthermore, very similar results were obtained by using molecular descriptors computed on 2D or 3D structures.

AB - Recent progress in the development of powerful tools suitable to design and to classify large chemical libraries can be fruitfully extended also to the ecotoxicity domain. Amidst these methods, Fuzzy Logic concepts, based on the possibility to handle the "concept of partial truth", constitute interesting approaches to developing general predictive models. In this work, a global strategy of Database Mining was applied on a data set of 568 chemicals, extracted from a toxicity database concerning the fathead minnow and divided into four classes, according to the toxicity ranges defined by the European Community legislation. Two large sets of molecular descriptors were tested on the 2D and 3D structures, and the best ones were selected with help of a procedure combining Genetic Algorithm concepts and stepwise method. After selecting the training set with a rational selection based on the Self Organizing Maps (SOM), structural-activity models were built by Adaptive Fuzzy Partition (AFP). This method consists in modeling relations between molecular descriptors and biological activities, by dynamically dividing the molecular descriptor hyperspace into a set of fuzzy subspaces. The best model was selected by a validation set, and its robustness was confirmed by predicting a test set of 80 chemicals never used to define the AFP models. An encouraging validation ratio of about 72% was obtained in the prediction of the experimental toxicity class. Furthermore, very similar results were obtained by using molecular descriptors computed on 2D or 3D structures.

KW - Ecotoxicity

KW - Fathead minnow

KW - Fuzzy logic

KW - SAR models

UR - http://www.scopus.com/inward/record.url?scp=0038745832&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038745832&partnerID=8YFLogxK

M3 - Article

VL - 22

SP - 210

EP - 219

JO - QSAR and Combinatorial Science

JF - QSAR and Combinatorial Science

SN - 1611-020X

IS - 2

ER -