![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081105122807im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
2007 Progress Report: Cheminformatics Tools for Toxicant Characterization
EPA Grant Number: R832721C004Subproject: this is subproject number 004 , established and managed by the Center Director under grant R832721
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
Center: New Jersey Research Center for Environmental Bioinformatics and Computational Toxicology
Center Director: Welsh, William J.
Title: Cheminformatics Tools for Toxicant Characterization
Investigators: Welsh, William J. , Androulakis, Ioannis , Floudas, Christodoulos , Georgopoulos, Panos G. , Ierapetritou, Marianthi , Rabitz, Herschel , Tong, Weida
Institution: Princeton University , Rutgers University , U.S. Food and Drug Administration , University of Medicine and Dentistry of New Jersey
EPA Project Officer: Mustra, David
Project Period: October 1, 2005 through September 30, 2010
Project Period Covered by this Report: October 1, 2006 through September 30, 2007
RFA: Computational Toxicology: Environmental Bioinformatics Research Center (2004)
Research Category: Computational Toxicology
Description:
Objective:The Research Center will bring together a team of computational scientists, with diverse backgrounds in bioinformatics, cheminformatics and enviroinformatics, from UMDNJ, Rutgers, and Princeton Universities, and the USFDA’s Center for Toxicoinformatics. This team will address, in a systematic and integrative manner, multiple elements of the toxicant Source-to-Outcome sequence (Investigational Area 1, as identified in the RFA) as well as develop cheminformatics tools for toxicant characterization (Investigational Area 2, Predictive Models for Hazard Identification). The computational tools to be developed through this effort will be extensively evaluated and refined through collaborative applications involving Center scientists as well as colleagues from the three universities and USEPA; particular emphasis will be on methods that enhance current quantitative risk assessment practices and reduce uncertainties.
Progress Summary:- The Shape Signatures tool has been adapted for applications in computational toxicology. Specifically, classification models were constructed for i) a series of ligands for the 5-hydroxytryptamine (serotonin) 2b receptor (5-HT2b receptor); and ii) a structurally diverse collection of inhibitors of the human Ether-a-go-go Related Gene (hERG) potassium channel. Three popular classification techniques were employed to build the models: k Nearest Neighbors (k-NN), Support Vector Machines (SVM), and Kohonen Maps (SOM). The accuracies of the Shape Signatures-SVM models for 5-HT2b (73-83%) and for hERG (69-73%) were found comparable to published results obtained using traditional QSAR molecular descriptors. These classification models for 5-HT2b represent the first global computational models for the cardiotoxicity associated with this receptor.
- A molecular modeling study has been initiated on the nuclear receptors (NRs) found in the liver, and their interactions with endogenous, natural and synthetic ligands including both agonists and antagonists. Computational pharmacophore models were derived for PXR agonists using separate data sets of imidazoles, steroids, and a set of diverse molecules for which published PXR agonist binding data are available. These pharmacophore models revealed that hydrophobic features are paramount for these agonists. In contrast, analysis of the corresponding PXR antagonist pharmacophore models based on biological data for azoles and biphenyls revealed that they are smaller with increased contributions by hydrophilic features. The azole antagonists (e.g., ketoconazole) were computationally docked in the proposed hydrophobic binding pocket of the AF-2 site. A combination of computational and experimental data for diverse classes of chemicals suggests that agonists and antagonists can bind at distinct regions on PXR. In particular, the findings reveal that certain azole and imidazole antagonists bind to the AF-2 site of PXR.
- A molecular modeling study was conducted on the structural and molecular basis of interactions of proteins with alkyl and aryl halides that are found in many commercial pesticides, disinfectants, and drugs. Statistical analysis of a database compiled from structural information found in the Protein Data Bank (PDB) revealed distinct patterns with respect to halogen interactions with specific types of atoms and groups in proteins.
- Endogenous ligands of nuclear hormone receptors tend to be highly conserved across species. In contrast, bile salts, the major elimination products of cholesterol, vary significantly across vertebrate species. Based on analysis of 1172 phylogenetically diverse vertebrates, it was found that the farnesoid X, vitamin D, and PXR receptors each acquired sensitivity to bile salts at different points in vertebrate evolution and adapted to evolutionary changes in bile salt structure and metabolism. Computationally structural models of the FXR receptor for various species played a key role in this analysis.
Currently planned and ongoing activities include the following:
- Continued implementation of existing and design of new ebTrack interfaces to open source databases (e.g. PostgreSQL) and to various “external” and Center-developed modeling tools for facilitating wider-deployment and applicability of the ebTrack/ArrayTrack system for integrative analyses of various types of genomic, proteomic, and metabonomic data. This will be pursued through further incorporation of novel, optimized statistical and pattern recognition methods for clustering of gene expression data as ebTrack components, and through further analysis of ongoing applications and initiation of additional applications of ArrayTrack for environmentally-relevant toxicants (e.g., dibutyl phthalate, Arsenic, etc.) and component-by-component evaluations of ArrayTrack applications.
- Refinement of the environmental bioinformatics Knowledge Base (ebKB) and making a public beta version of ebKB available.
- Continuing development and implementation of new techniques for incorporating biochemical data into the optimization and parameter estimation components of MENTOR-3P (Modeling Environment for Total Risk with Physiologically-based Pharmacokinetic modules for Populations), focusing on Bayesian tools in conjunction with optimization techniques.
- Refinements to the framework for DORIAN (Dose-Response Information Analysis) modules representing different scales of biological complexity ranging from molecule-molecule interactions to biochemical networks to virtual organs and systems.
- Implementation of a modular “Virtual Liver” with alternative levels of detail in describing physical structure of the liver with respect to toxicokinetic and toxicodynamic processes with case studies focusing on environmentally-relevant chemicals.
- Implementation of algorithms as DORIAN modules for rapid assessment of risks from chronic and multiscale exposures to mixtures of contaminants.
- Continuing development and incorporation of diagnostic tools as DORIAN modules for sensitivity and stability analysis of mechanistic models, and demonstration with case studies focusing on environmentally-relevant chemicals.
- Experimental verification of modeling results from network models of hepatocyte metabolism, and integration of regulatory rules within metabolic network models and constraining the model such that cell capabilities in the models become more realistic.
- Metabonomic case studies focusing on (a) pathways involved in steroidogenesis pathways due to in utero exposure to phthalate esters, (b) hepatocarcinogenic potential of exposure to triazole conazoles, and (c) experimental verification of the interactions between ethanol and other central hepatic pathways and xenobiotic pathways.
- Application of SNIP (S-space Network Identification Protocol) to larger networks with higher complexity and optimal design of perturbation experiments for improved efficiency and reliability of SNIP. Additional applications to other realistic bionetworks (e.g., metabolic networks) and optimizing the performance of SNIP.
- Further application of the RS-HDMR (Random-Sampling High Dimensional Model Representation) analysis of the mechanism of action on the cooperative inhibition of aspartate transcarbamoylase, which potentially can enable deeper understanding of many biological processes that this enzyme is involved in.
- Further incorporation of cheminformatic data in Shape Signatures classification models. Ongoing case studies focus on a blood-brain barrier model.
- Continuing development of structural models for liver nuclear receptors: PXR, FXR, LXR, VDR, etc., and, molecular modeling studies of xenobiotic-NR interactions, with emphasis on chemicals from the Toxcast database and nuclear receptors found in the liver. Ongoing case studies focus on the computational structural model of FXR for Ciona (sea squirt), for comparison with x-ray structural data of FXR for other species.
- Study of new approaches, including hybrid methods, for (a) de novo protein design, (b) understanding biological coherence in gene clustering, and (c) peptide identification.
- Definition of specific case studies for comprehensive source-to-outcome modeling and analysis, and further evaluation of approaches developed at the Center through collaborative efforts with external researchers.
No journal articles submitted with this report: View all 8 publications for this subproject
Supplemental Keywords:, ENVIRONMENTAL MANAGEMENT, Scientific Discipline, Health, Risk Assessment, Biology, Risk Assessments, Biochemistry, exposure assessment, biochemical research, chemical composition, ecological risk assessment, toxicologic assessment, bioinformatics, human health risk, biopollution, toxicology, environmental risks, risk, computational toxicology
Progress and Final Reports:
2006 Progress Report
Original Abstract
Main Center Abstract and Reports:
R832721 New Jersey Research Center for Environmental Bioinformatics and Computational Toxicology
Subprojects under this Center:
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
R832721C001 Development and Application of the DORIAN (Dose-Response Information Analysis) System
R832721C002 Hepatocyte Metabolism Model for Xenobiotics
R832721C003 Development of Computational Tools for Optimal Identification of Biological Networks
R832721C004 Cheminformatics Tools for Toxicant Characterization
R832721C005 Optimization Tools for In Silico Structural Proteomics