Science Inventory

Docking-based classification models for exploratory toxicology studies on high-quality estrogenic experimental data

Citation:

Trisciuzzi, D., D. Alberga, K. Mansouri, R. Judson, S. Cellamare, M. Catto, A. Carotti, E. Benfenati, E. Novellino, G. Mangiatordi, AND O. Nicolotti. Docking-based classification models for exploratory toxicology studies on high-quality estrogenic experimental data. Future Medicinal Chemistry. Future Science Group, London, Uk, 7(14):1921-1936, (2015).

Impact/Purpose:

Shows how structure-based methods, widely applied to drug discovery programs, can be adapted to meet the conditions of the regulatory context. Evaluation of 24 reliable docking-based classification models able to predict the estrogenic potential of a large collection of chemicals having high quality experimental data, kindly provided by the U.S. Environmental Protection Agency (EPA).

Description:

Background: Exploratory toxicology is a new emerging research area whose ultimate mission is that of protecting human health and environment from risks posed by chemicals. In this regard, the ethical and practical limitation of animal testing has encouraged the promotion of computational methods for the fast screening of huge collections of chemicals available on the market. Results: We derived 24 reliable docking-based classification models able to predict the estrogenic potential of a large collection of chemicals having high quality experimental data, kindly provided by the U.S. Environmental Protection Agency (EPA). The predictive power of our docking-based models was supported by values of AUC, EF1% (EFmax = 7.1), -LR (at SE = 0.75) and +LR (at SE = 0.25) ranging from 0.63 to 0.72, from 2.5 to 6.2, from 0.35 to 0.67 and from 2.05 to 9.84, respectively. In addition, external predictions were successfully made on some representative known estrogenic chemicals. Conclusion: We show how structure-based methods, widely applied to drug discovery programs, can be adapted to meet the conditions of the regulatory context. Importantly, these methods enable one to employ the physicochemical information contained in the X-ray solved biological target and to screen structurally-unrelated chemicals.

URLs/Downloads:

http://www.future-science.com/doi/abs/10.4155/fmc.15.103   Exit

Record Details:

Record Type: DOCUMENT (JOURNAL/PEER REVIEWED JOURNAL)
Product Published Date: 10/06/2015
Record Last Revised: 07/19/2016
OMB Category: Other
Record ID: 309932

Organization:

U.S. ENVIRONMENTAL PROTECTION AGENCY

OFFICE OF RESEARCH AND DEVELOPMENT

NATIONAL CENTER FOR COMPUTATIONAL TOXICOLOGY