4
R. Kusko and H. Hong
the intersection of computer science and toxicology is what we here call “Computational Toxicology.” Computational toxicology integrates both the long-standing
computational methods and the newer approaches including neural networks (NNs)
and artificial intelligence (AI). Rather than individual scientists and researchers trying
to understand multiple complex phenomena via bench experiments, these complex
biological systems can now be modeled and predicted in the computational space.
Issues which may have previously seemed impossible or intractable are increasingly
becoming solvable due to the scalability of computational toxicology.
1.2 Domain of Computational Toxicology
Safe drugs, safe food products, and a safe environment for living organisms are of
concern in all countries around the world. Toxicology leadership usually stems from
governmental regulatory agencies, in the USA including the FDA (Food and Drug
Administration), EPA (Environment Protection Agency), DA (Department of Agriculture), NCATS (National Center for Advancing Translational Sciences), NIEHS
(National Institute of Environmental Health Sciences) and others. These regulatory
agencies are responsible for maintaining the health and well-being of a population
and actively seek to prevent any exposures to toxic chemicals. Additionally, the pharmaceutical and biotechnology industry strive to improve patient lives by bringing
both new and generic medicines to the market and must do so while minimizing harm
to human life. Safety and toxicity screening is critical throughout the steps of any
drug development program, starting from the preclinical stage, during clinical trials,
and even in post-market surveillance. For the food and agriculture industry, safety
screening is also a key step in establishing safe exposure levels to new additives or
pesticides. Academics, while rarely developing a product for commercial purposes,
do seek to create and test toxicity screens and also assess toxicity mechanism of
action (MOA) or mode of action (MoA). This textbook emphasizes the methods of
computational toxicology and their potential applications in regulatory science, but
the topic is clearly relevant across sectors and around the world.
1.3 Need for Computational Toxicology
The field of computational toxicology has been blooming due to the fundamental limitations of experimental toxicology. While a dizzying array of novel chemical matter
is being created every day, traditional experiments are bottlenecked by throughput
and cost. In other words, the need for fast toxicity screening and prediction is ever
increasing and traditional in vivo and in vitro approaches cannot keep pace. Moreover, there is a global push to avoid the use of animals for experimental testing.
Traditional approaches are also limited in the number of doses, time points, organ
systems, and combinations that can possibly be tested sanely in one experiment or
laboratory.
R. Kusko and H. Hong
the intersection of computer science and toxicology is what we here call “Computational Toxicology.” Computational toxicology integrates both the long-standing
computational methods and the newer approaches including neural networks (NNs)
and artificial intelligence (AI). Rather than individual scientists and researchers trying
to understand multiple complex phenomena via bench experiments, these complex
biological systems can now be modeled and predicted in the computational space.
Issues which may have previously seemed impossible or intractable are increasingly
becoming solvable due to the scalability of computational toxicology.
1.2 Domain of Computational Toxicology
Safe drugs, safe food products, and a safe environment for living organisms are of
concern in all countries around the world. Toxicology leadership usually stems from
governmental regulatory agencies, in the USA including the FDA (Food and Drug
Administration), EPA (Environment Protection Agency), DA (Department of Agriculture), NCATS (National Center for Advancing Translational Sciences), NIEHS
(National Institute of Environmental Health Sciences) and others. These regulatory
agencies are responsible for maintaining the health and well-being of a population
and actively seek to prevent any exposures to toxic chemicals. Additionally, the pharmaceutical and biotechnology industry strive to improve patient lives by bringing
both new and generic medicines to the market and must do so while minimizing harm
to human life. Safety and toxicity screening is critical throughout the steps of any
drug development program, starting from the preclinical stage, during clinical trials,
and even in post-market surveillance. For the food and agriculture industry, safety
screening is also a key step in establishing safe exposure levels to new additives or
pesticides. Academics, while rarely developing a product for commercial purposes,
do seek to create and test toxicity screens and also assess toxicity mechanism of
action (MOA) or mode of action (MoA). This textbook emphasizes the methods of
computational toxicology and their potential applications in regulatory science, but
the topic is clearly relevant across sectors and around the world.
1.3 Need for Computational Toxicology
The field of computational toxicology has been blooming due to the fundamental limitations of experimental toxicology. While a dizzying array of novel chemical matter
is being created every day, traditional experiments are bottlenecked by throughput
and cost. In other words, the need for fast toxicity screening and prediction is ever
increasing and traditional in vivo and in vitro approaches cannot keep pace. Moreover, there is a global push to avoid the use of animals for experimental testing.
Traditional approaches are also limited in the number of doses, time points, organ
systems, and combinations that can possibly be tested sanely in one experiment or
laboratory.