> Q1] An mg-auto calculation is not suitable because I am
> forced to focus on a single point when doing the fine
> computation, am I right?
Right, but you could use something like parallel focusing to cover the
entire protein domain. We've recently added an asynchronous mode to APBS to
accomplish exactly this. It will be available in the next APBS release, due
out in a week or two. In the meantime, you could set up several separate
focusing calculations to cover the domain or run the calculation on a
> Q2] Checking the mg-manual examples (eg. PKA-LIG), I notice
> in the apbs1.in file that you do two calculations for each
> case: (mol 3 = complex)
> Then the Energy(Calculation 2.) is the right ligand Energy
> (i.e. the total electrostatic energy of all the ligand)
> not just the Energy of the region comprised in the grid
> of calculation 2, am I right?
Right; the energy calculation from 2 includes the energy from the portion of
the domain from calculation 1 not covered by the focused domain from 2.
> Q3] Would I have to follow the file
> to do what I want? Or the warning in the manual:
> "[...] it should not be used to look at subsets
> of biomolecules such as titration sites, etc.
> Such subset calculations require
> more complicated energy evaluation which [...]"
> applies, so... what do I have to do?
Yes, but it would be best to set the fine grid in the parallel focusing to
cover both proteins.
> Q4] I suppose I have to take into account solvation energies
> in my case, right? (I mean as in ex. FKBP:
Yes, or you could calculate it directly (not as safe) per the
apbs/examples/actin-dimer input files.