Improvement of quantitative structure–activity relationship (QSAR) tools for predicting Ames mutagenicity: outcomes of the Ames/QSAR International Challenge Project

Abstract

The International Conference on Harmonization (ICH) M7 guideline allows the use of in silico approaches for predicting Ames mutagenicity for the initial assessment of impurities in pharmaceuticals. This is the first international guideline that addresses the use of quantitative structure–activity relationship (QSAR) models in lieu of actual toxicological studies for human health assessment. Therefore, QSAR models for Ames mutagenicity now require higher predictive power for identifying mutagenic chemicals. To increase the predictive power of QSAR models, larger experimental datasets from reliable sources are required. The Division of Genetics and Mutagenesis, National Institute of Health Sciences (DGM/NIHS) of Japan recently established a unique proprietary Ames mutagenicity database containing 12140 new chemicals that have not been previously used for developing QSAR models. The DGM/NIHS provided this Ames database to QSAR vendors to validate and improve their QSAR tools. The Ames/QSAR International Challenge Project was initiated in 2014 with 12 QSAR vendors testing 17 QSAR tools against these compounds in three phases. We now present the final results. All tools were considerably improved by participation in this project. Most tools achieved >50% sensitivity (positive prediction among all Ames positives) and predictive power (accuracy) was as high as 80%, almost equivalent to the inter-laboratory reproducibility of Ames tests. To further increase the predictive power of QSAR tools, accumulation of additional Ames test data is required as well as re-evaluation of some previous Ames test results. Indeed, some Ames-positive or Ames-negative chemicals may have previously been incorrectly classified because of methodological weakness, resulting in false-positive or false-negative predictions by QSAR tools. These incorrect data hamper prediction and are a source of noise in the development of QSAR models. It is thus essential to establish a large benchmark database consisting only of well-validated Ames test results to build more accurate QSAR models.

Identifying Compounds with Genotoxicity Potential Using Tox21 High-Throughput Screening Assays

“…Thus, there is a demonstrated need for alternative methods to characterize compound genotoxicity and/or to prioritize the thousands of inadequately tested compounds for evaluation in traditional genotoxicity assays, to better focus available resources where the need is greatest. One such alternative method is the use of in silico quantitative structure-activity relationship (QSAR) models for predicting bacterial mutagenicity 4 . However, genotoxicity is not limited to the induction of point mutations in bacteria, but also includes other types of genetic damage including chromosomal changes (both numerical and structural), and DNA damage (e.g., DNA adducts, DNA-DNA crosslinks).…”

An assessment of mutagenicity of chemical substances by (quantitative) structure–activity relationship

“…As the result of three trials of the Ames/QSAR International Challenge Project, all QSAR tools were considerably improved. Most tools achieved > 50% sensitivity and accuracy was as high as 80%, which is almost equivalent to the inter-laboratory reproducibility of Ames tests (Table 7), implying that the project was successfully completed [23]. The DGM/NIHS will start the next Ames/QSAR International Challenge Project near future, because more than 2000 new chemicals' Ames tests results submitted to ANEI-HOU has been accumulated during recent a few years.…”

In silico toxicology: From structure–activity relationships towards deep learning and adverse outcome pathways

“…However, they concluded that the evaluated models are not yet predictive enough to use them as a stand‐alone tool . In contrast, a large international study revealed that existing and commonly used models for Ames mutagenicity are working well and the predictivity can compete with the in vitro Ames test . In this context it has to be noted that, depending on the dataset used, it might be that the contaminants Frenzel and coworkers used are too specialized to be caught by models in the public domain, which usually also only rely on public data.…”

“…Thus, their molecules might not be in the applicability domain (see also Section 4.5). Another important finding by Honma and coworkers was that incorporating newly available data also increases the performance and thus models should be updated regularly …”