Although electron microscopy (EM) allows determining the structure of macromolecular complexes at near-atomic resolution, high resolution is still difficult to achieve for complexes that are flexible, heterogenous, or imaged in cells with cryo-electron tomography. A state-of-the-art methodology to obtain structures of such complexes is integrative structural modelling, which combines data from complementary techniques, such as X-ray crystallography, EM, NMR, SAXS or cross-linking mass spectrometry. The current integrative modelling software, however, is either limited to complexes of simple architectures or requires custom modifications, and thus a major expertise, when applied to more complex architectures. Thus, we developed a versatile modelling pipeline for integrative modelling of protein complexes with very complex architecture in an accessible and efficient way. Our pipeline is implemented using Integrative Modeling Platform library, on top of which it implements custom algorithms for efficient sampling of the conformational space, versatile configuration language and graphical interface for input preparation, modelling and analysis, and additional restraints. Flexible and symmetrical modeling is also supported. The pipeline enables modeling of very complex architectures, for example, with multiple symmetries or with EM maps for alternative conformational states. I will present the pipeline and how we applied it to model complexes ranging from the 1 MDa Elongator to the 110 MDa human nuclear pore complex (NPC). For the Elongator complex, I will present how our published model agrees with the subsequent high-resolution cryo-EM structure. For the NPC, I will show our most recent results of how the pipeline allowed building structural models of NPCs from yeast S. cerevisiae and S. pombe, including a novel multi-state modelling protocol for modelling based on multiple EM maps simultaneously. The new NPC models reveal surprisingly different architectures and depict conformational changes of NPCs in response to cellular states.