In this document the importance of mechanistic consideration in quantitative structure-activity relationship ((Q)SAR) analysis, the critical role of mechanistic consideration in improving various (Q)SAR approaches and possible integrative approaches of combining chemoinformatics and bioinformatics are discussed, mainly using carcinogenicity as an illustrative toxicity endpoint. The principles and issues described in this document are general and may also be used for various types of chemical assessment. The intended target of readership is for global chemical hazard/risk assessors in regulatory agencies, industries, non-governmental organization and academia who require reliable and scientifically supportable (Q)SAR information and predictions in their assessments as well as for developers/researchers who endeavour to produce scientifically reliable (Q)SAR predictive models and tools.