This command calculates and selects new distance restraints. See Restraints.make() for full details.
Distance restraints are generated for all pairs of atoms where atom is from selection atmsel1 and atom is from selection atmsel2. Moreover, for a restraint to be created, at least one distance in the template structures must be less than maximal_distance (in Å). The mean of this basis pdf is equal to the template distance and its standard deviation is calculated from an analytic model specified by distance_rsr_model. Use model 5 for Cα- Cα distances and model 6 for N-O distances. For models 1 through 6, this standard deviation is transformed by where and are given by restraint_stdev and is a gap weighting function of the form . is the average distance of the two residues involved in the restraint from a gap. For models 3 through 6, this is additionally transformed by where the sum is over each of the atoms involved in the distance, and are given by restraint_stdev2, and is the distance of each residue from a gap. The first six models are polynomials and depend on several structural features of the template and its similarity to the target. The polynomial coefficients are specified in library file '$PARAMS_LIB'. When “polynomial model” 7 is selected, the standard deviation of restraints is set to constant . Each basis pdf in the distance pdf corresponds to one template structure with an equivalent distance.
In addition, the atom pairs restrained by homology-derived restraints must by default not be in a chemical bond, chemical angle, dihedral angle, or on an excluded pairs list. This behavior can be changed by resetting energy_data.excl_local (see conjugate_gradients()).
If the restrained distance for a given atom pair is less than exclude_distance, that pair is also excluded from the nonbonded list. This is useful if you are building short distance restraints to approximate bonds.
An 'equivalent' distance is defined as the distance between a pair of equivalent atoms. See Atom.get_equivalent_atom() for the rules for determining equivalency.