Intrinsically disordered proteins are biomolecules that play important roles in many cellular func- tions, and their dysfunction is often associated with common human diseases. For this reason, there is considerable interest in understanding how these molecules function in atomistic detail. However, unlike globular proteins, intrinsically disordered proteins cannot fold into stable con- formations. Rather, they can adopt an exorbitant number of unfolded conformations that are rapidly interchangeable. This structural heterogeneity makes it difficult to study intrinsically dis- ordered proteins and to characterize their structural ensembles at atomistic resolution. In fact, the characterization of these molecules often requires an interplay between multiple experiments, that probe sparse structural information associated with their conformational ensembles in solution, and computational modelling, that exploits the experiments to reconstruct accurate conforma- tional ensembles at atomistic resolution. This framework combining experiments and modelling applied to biological problems is often referred to as “integrative structural biology”, or in a more general context as “integrative modeling”. This thesis deals with several aspects related to the inte- gration of experimental data (mainly from small angle X-ray scattering and NMR) and molecular simulations of intrinsically disordered proteins. Various approaches are used to sample the vast landscape of conformations that intrinsically disordered proteins can adopt. Special emphasis is placed on forward models, i.e., the algorithms used to compute experimental observables from conformational ensembles, and on methods to improve the agreement between experiments and structural models. Part of the thesis is devoted to cases that illustrate the importance of using multiple and independent sources of experimental information to obtain accurate models of in- trinsically disordered proteins, as well as reliable forward models. While most of the thesis deals with computational modelling as the second step of the integrative process, the last part of the thesis presents an opposite scenario. In this part, many concepts and methods from integrative modelling are applied to the development of an algorithm for the design of intrinsically disor- dered proteins. This algorithm is used to investigate the relationship between the arrangement of amino acids in the sequence of intrinsically disordered proteins, their structural features and their propensity to form biomolecular condensates. Finally, some of the designed sequences are experimentally validated.