IMP SETUP repeat(imp-setup-options) atom-selection imp-setup-options ::= [SOLVent real] [INTErior real] [MAXIter integer] [OUTIter integer] [TOLErance real] [radius-edit-options] [charge-edit-options] [KEEP] [very-bogus-experimental-options] radius-edit-options ::= repeat([RADIus edit-type real atom-selection END]) charge-edit-options ::= repeat([CHARge edit-type real atom-selection END]) edit-type ::= [SET] [SCALe] [SHIFt] very-bogus-experimental-options ::= [VOLUme volume-option SCALe real] volume-options ::= [LINEar] [QUADratic] [CUBIc] [POWEr]
See Atom Selection, for the syntax of atom-selection.
The IMP SETUP command initializes the data structure in preparation for solving for the induced multipole moments.
The environment can be defined. SOLVent real sets the solvent dielectric constant (default 78). INTErior real sets the solute interior dielectric constant (default 2).
The iteration control can be set (although this can be postponed or reset on the IMP SOLVE command line). MAXIter integer sets the maximum number of iterations (default 20). OUTIter integer prints the convergence status every integer iterations. Zero, the default value, indicates that the convergence status should only be printed at termination. TOLErance real sets the fractional change in the total energy below which iteration will be terminated to real (default 1.E-06).
Atomic parameters can be modified. Multiple RADIus and CHARge options may be used to modify the atomic radii and charges of selected groups of atoms. The modifications include setting them to a value with SET, multiplying them by a some value with SCALe, and adding a particular value to them with SHIFT. These effects of these options are cumulative and applied in the order they appear on the command line. So, for example, RADIUS SCALE 2 END RADIUS SHIFT -1 END would change all radii to twice their original values minus one Angstrom.
The KEEP option requests that the Coulombic tensors be saved between iterations. This saves considerable calculation time, but requires storage of 7*NATOM*(NATOM-1) additional reals.
The "very-bogus-experimental-options" are all part of ongoing efforts to improve the method. These will disappear even before IMPS rises from being listed in the experimental section.