Upper panel: Schematic domain structures of six SASSIE-modelled macromolecules. The domains are drawn approximately to scale according to their molecular structures. The major linkers varied in SASSIE searches are denoted by arrows (Table 1); some simplification has been made. The domain names are as follows: HIV-1 Gag: MA, matrix; CA, capsid; NC, nucleocapsid. MASP: CUB, C1r/C1s, Uefg and bone morphogenetic protein-1; EGF, epidermal growth factor; SCR, short complement regulator; SP, serine protease. Ub, ubiquitin. IgG2 and IgA1: Fab, fragment antigen binding; Fc, fragment crystallizable. Lower panel: Molecular structures for these six macromolecules, all drawn to the same scale in PYMOL (Schrodinger LLC). The best-fit structures are to be described as ensembles of structures and not as the single structures as shown. The domain colours follow those in the upper panel. That for HIV-1 Gag is taken from the starting model for the simulations, where the MA, CA and NC domains are taken from crystal or NMR structures, and the p2 domain is not shown (Table 1). That for MASP-3D is taken from the crystal structure of MASP-1 and was the starting model used to initiate the fitting. That for the K27-ubiquitin dimer is taken from the isopeptide dimer formed through Lys27 (distal Ub, orange; proximal Ub, yellow; K27, magenta). That for IgA1 is the final model from the SASSIE fits, but not showing the glycan chains (Fig. 4). That for the Hfq–mRNA complex is the input file used in the Complex MC tutorial; under this is the starting crystal structure of the Hfq core protein bound with two heptamer nucleotide chains (PDB code 4ht8). The mRNA chains are shown in orange.