I am currently modelling a system with a fair amount of restraints
(symmetry, secondary structure, and distance) that takes over an hour to
model. I have found that if i make a mistake in naming one of the
residues/atoms in one of my restraints, that the program does not catch
this right away, but will crash after about an hour when it finally
notices that the restraint is invalid. This is a pretty inconvenient
way in realizing my input mistakes, and am wondering if there is a way
to have Modeller check that all of my restraints are valid before the
program begins processing the models.
Modeller cannot check whether the restraints are valid until it knows
how many atoms are in your system, so you're always going to have to
wait for the initial model to be generated. But your custom restraints
are applied anyway after that model is generated, after the
homology-derived restraints, so are you saying that the generation of
the initial model and homology-derived restraints takes an hour? That
seems excessive.