The aim of this theoretical study is to characterize the relationship between the structure and the physicochemical properties of zinc oxide nanoparticles (ZnO NPs) and Mn doped ZnO NPs to assess the toxicological impact of these materials. In order to do so, a multiscale modelling approach is applied. Different nanoparticles – in terms of geometry, aggregation and size – are characterized by their electronic properties using quantum mechanics methodologies (i.e., DFT, Density Functional Theory). To evaluate the toxicology impact of ZnO NPs in human health safety, the interaction between the material and biological target systems are modelled. Applying classical molecular dynamics methodologies, the interaction between a set of selected human proteins with the NPs are tested. Based on the Adverse Outcome Pathway (AOP) concept, identifying the Molecular Initiating Event (MIE) – the first event that triggers the myriad of events that cause the adverse outcome – allow to develop Safe by Design materials. In this scheme, the interaction of the studied nanoparticles with model human cell membranes is proposed as a pivotal MIE, which is studied through Molecular Dynamics Simulations for different ZnO NPs in terms of shape, size and concentration. The interaction between the NPs and model cell membranes is quantified in terms of membrane nanodescriptors which are scaled up through Machine Learning–based methods for developing simple and predictive tools. This work is included as a case study for the Horizon - DIAGONAL project which aims to develop Sustainable and Safe By Design (SSbDs) materials, process and technologies to guarantee the long-term nanosafety for Multicomponent Nanomaterials (MCNMs) and High Aspect Ratio Nanoparticles (HARNs) in their whole life cycle