Figure 4: PanDDA maps reveal complex minor conformations and identify allosteric binders.

In JMJD2D data set x402, at 1.45 Å, (a) conventional maps (contoured as in Fig. 3a) show a complex superposition of bound and unbound states that make it impossible to identify the bound state (the known unbound state is shown). (b) However, in PanDDA maps (contoured as in Fig. 3c, BDC=0.8) the bound conformation can be modelled easily (as shown). (c) Final models for the unbound (yellow) and bound (magenta) conformations highlight the large conformational change. (d) Fragments are detected to bind all over the surface of JMJD2D, revealing potential allosteric sites, including the peptide-binding groove (site A) and the large helix reordering (site B).