Predictive Carcinogenicity

A Model for Aromatic Compounds, with Nitrogen-Containing Substituents, Based on Molecular Descriptors Using an Artificial Neural Network

Giuseppina Gini, Marco Lorenzini, Emilio Benfenati, Paola Grasso, Maurizio Bruschi

Research output: Contribution to journalArticle

57 Citations (Scopus)

Abstract

A back-propagation neural network to predict the carcinogenicity of aromatic nitrogen compounds was developed. The inputs were molecular descriptors of different types: electrostatic, topological, quantumchemical, physicochemical, etc. For the output the index TD50 as introduced by Gold and colleagues was used, giving a continuous numerical parameter expressing carcinogenicity. From the tens of descriptors calculated, principal component analysis enabled us to restrict the number of parameters to be used for the artificial neural network (ANN). We used 104 molecules for the study. An Rcv2 = 0.69 was obtained. After removal of 12 outliers, a new ANN gave an Rcv2 of 0.82.

Original languageEnglish
Pages (from-to)1076-1080
Number of pages5
JournalJournal of Chemical Information and Computer Sciences
Volume39
Issue number6
Publication statusPublished - 1999

Fingerprint

Aromatic compounds
Nitrogen
Neural networks
Nitrogen Compounds
Nitrogen compounds
Backpropagation
Gold
Principal component analysis
Electrostatics
Molecules

ASJC Scopus subject areas

  • Chemistry(all)
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Information Systems

Cite this

Predictive Carcinogenicity : A Model for Aromatic Compounds, with Nitrogen-Containing Substituents, Based on Molecular Descriptors Using an Artificial Neural Network. / Gini, Giuseppina; Lorenzini, Marco; Benfenati, Emilio; Grasso, Paola; Bruschi, Maurizio.

In: Journal of Chemical Information and Computer Sciences, Vol. 39, No. 6, 1999, p. 1076-1080.

Research output: Contribution to journalArticle

@article{d5ed9cd5f30e410aacd40bcac26259bb,
title = "Predictive Carcinogenicity: A Model for Aromatic Compounds, with Nitrogen-Containing Substituents, Based on Molecular Descriptors Using an Artificial Neural Network",
abstract = "A back-propagation neural network to predict the carcinogenicity of aromatic nitrogen compounds was developed. The inputs were molecular descriptors of different types: electrostatic, topological, quantumchemical, physicochemical, etc. For the output the index TD50 as introduced by Gold and colleagues was used, giving a continuous numerical parameter expressing carcinogenicity. From the tens of descriptors calculated, principal component analysis enabled us to restrict the number of parameters to be used for the artificial neural network (ANN). We used 104 molecules for the study. An Rcv2 = 0.69 was obtained. After removal of 12 outliers, a new ANN gave an Rcv2 of 0.82.",
author = "Giuseppina Gini and Marco Lorenzini and Emilio Benfenati and Paola Grasso and Maurizio Bruschi",
year = "1999",
language = "English",
volume = "39",
pages = "1076--1080",
journal = "Journal of Chemical Information and Computer Sciences",
issn = "0095-2338",
publisher = "American Chemical Society",
number = "6",

}

TY - JOUR

T1 - Predictive Carcinogenicity

T2 - A Model for Aromatic Compounds, with Nitrogen-Containing Substituents, Based on Molecular Descriptors Using an Artificial Neural Network

AU - Gini, Giuseppina

AU - Lorenzini, Marco

AU - Benfenati, Emilio

AU - Grasso, Paola

AU - Bruschi, Maurizio

PY - 1999

Y1 - 1999

N2 - A back-propagation neural network to predict the carcinogenicity of aromatic nitrogen compounds was developed. The inputs were molecular descriptors of different types: electrostatic, topological, quantumchemical, physicochemical, etc. For the output the index TD50 as introduced by Gold and colleagues was used, giving a continuous numerical parameter expressing carcinogenicity. From the tens of descriptors calculated, principal component analysis enabled us to restrict the number of parameters to be used for the artificial neural network (ANN). We used 104 molecules for the study. An Rcv2 = 0.69 was obtained. After removal of 12 outliers, a new ANN gave an Rcv2 of 0.82.

AB - A back-propagation neural network to predict the carcinogenicity of aromatic nitrogen compounds was developed. The inputs were molecular descriptors of different types: electrostatic, topological, quantumchemical, physicochemical, etc. For the output the index TD50 as introduced by Gold and colleagues was used, giving a continuous numerical parameter expressing carcinogenicity. From the tens of descriptors calculated, principal component analysis enabled us to restrict the number of parameters to be used for the artificial neural network (ANN). We used 104 molecules for the study. An Rcv2 = 0.69 was obtained. After removal of 12 outliers, a new ANN gave an Rcv2 of 0.82.

UR - http://www.scopus.com/inward/record.url?scp=0033217369&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033217369&partnerID=8YFLogxK

M3 - Article

VL - 39

SP - 1076

EP - 1080

JO - Journal of Chemical Information and Computer Sciences

JF - Journal of Chemical Information and Computer Sciences

SN - 0095-2338

IS - 6

ER -