1 Computational Toxicology Promotes Regulatory Science
7
software tools for MD simulations and the challenges to apply these software tools
to computational toxicology and summarized key protocols to run MD simulations.
The applicability domain of a prediction model is defined as the structural space
that is covered by the chemicals of the training set. It is expected that the predictions
from the model for new compounds within the structural space are more accurate than
the predictions of chemicals out of the space. Analysis of applicability domains in
computational toxicology is important for assessing QSAR models [28]. Chapter 11
reviewed different perspectives of the applicability domain and the existing methods for analysis of applicability domain. It also formalized a holistic approach for
utilization of the applicability domain in computational toxicology.
1.5 Potential Applications of Computational Toxicology
in Regulatory Science
Computational toxicology has been accepted in the regulation of products. One of
the examples is the International Council for Harmonisation M7 (ICH M7) guideline that describes the assessment of carcinogenic risk of mutagenic impurities in
drug products [29]. This indicates the state of the art of a computational toxicology
method and is the milestone for regulatory acceptance of computational toxicology
for pharmaceutical products [30, 31]. In the USA, the FDA accepted QSAR modeling results for impurities in applications of drug products. The FDA developed the
Medical Device Development Tools (MDDT) program to qualify tools that can be
used in evaluation of medical devices [32]. In the newly released FDA’ predictive
toxicology roadmap, computational toxicology is listed as one of the new technologies might be able to address some of the needs in regulatory science [33]. The EPA’s
Endocrine Disruptor Screening Program (EDSP) in the twenty-first century is using
computational toxicology, coupling with in vitro methodologies, to prioritize and
identify EDSP Tier 1 information needs for pesticide active ingredients that will be
included in the registration review program [34]. In Europe, read-across, a commonly
used computational toxicology method, is adopted for data gap filling in registrations
submitted under the REACH regulation [35]. Computational toxicology is gaining
attention in chemical risk assessment and management in China [36]. This book’s
solicited chapters shed lights on examples of potential applications of computational
toxicology in regulatory science in USA, Europe, and China.
In terms of consumer food safety, toxicokinetics, QSAR modeling, and bioinformatics approaches are currently in use. Over time, certainly many more approaches
will be added to screen for toxic food products. Chapter 12 reviewed quantitative
structure–activity relationships, toxicokinetic modeling and simulation, and bioinformatics in the FDA’s Center for Food Safety and Applied Nutrition in-house food
ingredient knowledgebase to show the scientific utility of computational toxicology
for improving regulatory review efficiency.
7
software tools for MD simulations and the challenges to apply these software tools
to computational toxicology and summarized key protocols to run MD simulations.
The applicability domain of a prediction model is defined as the structural space
that is covered by the chemicals of the training set. It is expected that the predictions
from the model for new compounds within the structural space are more accurate than
the predictions of chemicals out of the space. Analysis of applicability domains in
computational toxicology is important for assessing QSAR models [28]. Chapter 11
reviewed different perspectives of the applicability domain and the existing methods for analysis of applicability domain. It also formalized a holistic approach for
utilization of the applicability domain in computational toxicology.
1.5 Potential Applications of Computational Toxicology
in Regulatory Science
Computational toxicology has been accepted in the regulation of products. One of
the examples is the International Council for Harmonisation M7 (ICH M7) guideline that describes the assessment of carcinogenic risk of mutagenic impurities in
drug products [29]. This indicates the state of the art of a computational toxicology
method and is the milestone for regulatory acceptance of computational toxicology
for pharmaceutical products [30, 31]. In the USA, the FDA accepted QSAR modeling results for impurities in applications of drug products. The FDA developed the
Medical Device Development Tools (MDDT) program to qualify tools that can be
used in evaluation of medical devices [32]. In the newly released FDA’ predictive
toxicology roadmap, computational toxicology is listed as one of the new technologies might be able to address some of the needs in regulatory science [33]. The EPA’s
Endocrine Disruptor Screening Program (EDSP) in the twenty-first century is using
computational toxicology, coupling with in vitro methodologies, to prioritize and
identify EDSP Tier 1 information needs for pesticide active ingredients that will be
included in the registration review program [34]. In Europe, read-across, a commonly
used computational toxicology method, is adopted for data gap filling in registrations
submitted under the REACH regulation [35]. Computational toxicology is gaining
attention in chemical risk assessment and management in China [36]. This book’s
solicited chapters shed lights on examples of potential applications of computational
toxicology in regulatory science in USA, Europe, and China.
In terms of consumer food safety, toxicokinetics, QSAR modeling, and bioinformatics approaches are currently in use. Over time, certainly many more approaches
will be added to screen for toxic food products. Chapter 12 reviewed quantitative
structure–activity relationships, toxicokinetic modeling and simulation, and bioinformatics in the FDA’s Center for Food Safety and Applied Nutrition in-house food
ingredient knowledgebase to show the scientific utility of computational toxicology
for improving regulatory review efficiency.