Superfund Research Program
Computational Modeling of Mammalian Biomolecular Responses Core
Project Leader: Melvin E. Andersen (ScitoVation LLC)
Grant Number: P42ES004911
Funding Period: 2006-2020
Project Summary (2006-2013)
Knowledge of the shape of the dose-response curve must extend to levels at which humans are typically exposed if we are to accurately assess the risks of adverse effects on the public health from exposures to environmental chemicals. Health effects data are usually sparse at environmental levels of exposure and computational models are being used to estimate both chemical disposition (i.e., pharmacokinetics) and tissue responses (i.e., pharmacodynamics). While current pharmacokinetic models incorporate physiological and anatomical information to provide accurate estimates of target tissue doses, the pharmacodynamic relationship between a chemical at its target site and the ultimate biological effect is usually described empirically or semi-empirically. Molecular level descriptions of pharmacodynamic mechanisms would provide a better understanding of dose-response curves and would reduce uncertainty in safety and risk assessments. The mission of this Core is to provide the skills and resources needed to develop computational models of biochemical pathways and to thereby provide insight into the adverse health effects of TCDD and related chemicals. Since development of computational models is an iterative process, with model development and laboratory experiments proceeding hand-in-hand, this work is collaborative with the work in the Research Projects that the Core supports. The overall approach used for development of computational models is defined by 4 tasks: (1) Develop initial descriptions of biochemical pathways where the nodes of the pathway and the interactions between nodes are linked to biomedical databases; (2) Develop a directed graph by curating the pathway description obtained under (1); (3) Develop computational models based on the network structures described by directed graphs; (4) Determine if a stochastic or Boolean model is preferable to an ODE-based model for understanding the dynamic behavior of a particular biochemical network. Core staff also seek to train postdoctoral fellows and other staff from the Research Projects in the use of software for development of pathway maps and for computational modeling of the pathways.