Integrative structure modeling is often used to characterize structures and dynamics of large macromolecular assemblies by relying on multiple types of input information. The individual proteins or domains are represented by atomic resolution structures or low-resolution sphere models and data from a variety of sources, such as cross-linking mass spectrometry, cryo-Electron Microscopy, Small Angle x-ray scattering is used to assemble the subunits. Recent progress in protein folding enabled by deep learning by AlphaFold2 and RosettaFold provided an improved structural coverage for domains, and even protein-protein interactions used in Integrative Structure Modeling. However, these methods depend on multiple sequence alignment, that is not available for immune response complexes, such as antibody-antigen interactions. Recently, we began utilizing deep learning approaches for a range of integrative modeling tasks, including development of scoring functions for prediction of protein-protein or protein-peptide interactions, modeling and docking of antibodies and nanobodies to the antigens, binding sites identification, and learning scoring functions for experimental data. I will describe the progress and current challenges in deep learning applications for modeling of complexes.