What is Computational Toxicology?

Computational Toxicology as a Discipline

Computational toxicology is a multi-disciplinary science that requires understanding of:

  • Computational sciences
  • Systems biology
  • Exposure science
  • Risk assessment and risk management

Provided there is a sufficient database of information, almost any toxicological endpoint can be modelled using computational toxicology software. The approaches used and the endpoints analyzed are as diverse as the users employing these methods. Examples of these include pathway, systems and tissue modelling, cheminformatics, big data methodologies for analysis of information, screening of chemicals, and uncertainty analysis. In addition, computational models have been used to calculate potential toxicokinetic and pharmacokinetic parameters.

At present, there is regulatory acceptance of the use of computational toxicology models in replacement of the reverse bacterial mutation (Ames) assay for impurities in pharmaceuticals, endrogen activity of environmental chemicals, and read-across for a variety of endpoints.

The Role of Computational Toxicology in the Future

Broadly, the results of toxicity tests in non-human systems are used to predict what may occur in humans. This standard for testing is associated with uncertainties and, in some cases, lack of success in extrapolating data from such systems to humans, particularly in patient groups and consumers.

Currently, the lead test system classes for safety evaluation are in vitro and in vivo tests, both of which have flaws and problems in translation of their output to humans or other species. In addition, the use of animals in toxicity testing is coming under increased pressure, and although their use is unlikely to end in the foreseeable future, it is clear that the use of animals will diminish as they come to be used in a more focused and targeted manner.

It is generally accepted that the use of two systems can elucidate the human relevance of toxicological data and that this is likely to be enhanced by judicious use of computational (in silico) toxicology. It is possible to foresee a paradigm for safety evaluation and risk assessment that increases the emphasis placed on in vitro and in silico data, while reducing the role of tests in vivo. Use of computational methods prior to in vivo testing serves as an early screening tool for compounds with potentially undesirable characteristics, which can lead to a further reduction in unnecessary animal usage.

It is very likely that computational toxicology will become one of three test arenas for the development of new chemicals, including pharmaceuticals, in which a combination of the following systems is used to predict safety and aid risk assessment:

  • Computational toxicology in silico
  • In vitro assessment
  • In vivo tests to confirm predictions