The question was: "Can anyone tell me how to do move_fragment on my ligand ? My inhibitor atoms don't move; O always says:

 "IP4 (my inhibitor name) has no dictionary".

Could somebody teach me how to do this ?"

I wouldn't use move_fragment unless all else fails (the same is true for move_atom). Try the following instead:

• move_zone to roughly position the ligand in the density
• rsr_rigid to optimise fit between ligand and map
• tor_general to twist unconstrained dihedrals
• rsr_rigid for the finishing touch

in this fashion, you keep good bond lengths, angles and most dihedrals plus the stereo-chemistry will be retained (with move_fragment you could easily change an R into an S stereo-isomer without intending to do so !)

Alwyn added: "O has a database that describes how atoms within a particular residue type are too be treated by move_frag. If the residue type is AXR it might look like the following

 rsr_dict_axr t 3 70
 c1 c2 c3
 c4 c5 c6
 c7 c8 c9

clicking on C5 would then let you move the 3 atoms C4,C5,C6. Look at the distributed rsr_dict.o files"