Prediction of acute toxicity to fathead minnow by local model based QSAR and global QSAR approaches

Youngyong In, Sung Kwang Lee, Pil Je Kim, Kyoung Tai No

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h LC50 (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients (R2) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity (R2=0.663) on the test set.

Original languageEnglish
Pages (from-to)613-619
Number of pages7
JournalBulletin of the Korean Chemical Society
Volume33
Issue number2
DOIs
Publication statusPublished - 2012 Feb 20

Fingerprint

Toxicity
Linear regression
Neural networks
Chemical compounds
Learning systems

All Science Journal Classification (ASJC) codes

  • Chemistry(all)

Cite this

@article{5dd4e3d9eba84bab83e6c5b63f59daff,
title = "Prediction of acute toxicity to fathead minnow by local model based QSAR and global QSAR approaches",
abstract = "We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h LC50 (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients (R2) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity (R2=0.663) on the test set.",
author = "Youngyong In and Lee, {Sung Kwang} and Kim, {Pil Je} and No, {Kyoung Tai}",
year = "2012",
month = "2",
day = "20",
doi = "10.5012/bkcs.2012.33.2.613",
language = "English",
volume = "33",
pages = "613--619",
journal = "Bulletin of the Korean Chemical Society",
issn = "0253-2964",
publisher = "Korean Chemical Society",
number = "2",

}

Prediction of acute toxicity to fathead minnow by local model based QSAR and global QSAR approaches. / In, Youngyong; Lee, Sung Kwang; Kim, Pil Je; No, Kyoung Tai.

In: Bulletin of the Korean Chemical Society, Vol. 33, No. 2, 20.02.2012, p. 613-619.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Prediction of acute toxicity to fathead minnow by local model based QSAR and global QSAR approaches

AU - In, Youngyong

AU - Lee, Sung Kwang

AU - Kim, Pil Je

AU - No, Kyoung Tai

PY - 2012/2/20

Y1 - 2012/2/20

N2 - We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h LC50 (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients (R2) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity (R2=0.663) on the test set.

AB - We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h LC50 (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients (R2) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity (R2=0.663) on the test set.

UR - http://www.scopus.com/inward/record.url?scp=84863147203&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863147203&partnerID=8YFLogxK

U2 - 10.5012/bkcs.2012.33.2.613

DO - 10.5012/bkcs.2012.33.2.613

M3 - Article

AN - SCOPUS:84863147203

VL - 33

SP - 613

EP - 619

JO - Bulletin of the Korean Chemical Society

JF - Bulletin of the Korean Chemical Society

SN - 0253-2964

IS - 2

ER -