Dr. Robert Foster, Lhasa Limited Senior Scientist presented these slides within a 2020 Webinar; ‘ICH M7 Expert Review Workshop: Resolving common prediction scenarios using automated arguments in Nexus 2.3’.
The use of in silico predictions for the assessment of bacterial mutagenicity under the ICH M7 guideline is recommended when two complementary (Q)SAR models are used. Using two systems may increase the sensitivity and accuracy of predictions but also increases the need to review predictions, particularly in situations where results disagree. Expert review may be as simple as a cursory check that two positive predictions seem appropriate or it may be a full assessment of the chemical and biological activity of a compound and the training set examples. More often than not expert review is a relatively straight forward process and repetitive arguments may be used to resolve commonly occurring prediction scenarios. With this in mind, the ICH M7 function in Nexus 2.3 was updated to present the user with automated arguments that are generated based on the specific Derek and Sarah predictions. Presenting the user with automated arguments is intended to make expert review a more consistent and time-efficient process. This workshop presented the new features of Nexus 2.3 through demonstrating their utility to expert review of selected examples that represent common prediction scenarios.