Is is possible in Modeller to fully optimize the specified loop but also
to allow a little bit of "adjustment freedom" to the loop end residues
(here 199 and 206)? If so, how should I modify the script?
The residues you select as the loop are optimized, and the remainder of
the protein is not moved at all, so there is no real way to "half
select" terminal residues in the way you propose. But there are a few
things you could try. For example, you could extend the loop selection
to include residues 199 and 206, but add additional position restraints
on some or all atoms in those two residues to prevent them from moving
too much. Alternatively, you could select only some of the atoms (for
example, just the sidechain and part of the backbone) from those two
residues.
Problem 2:
Could anyone advise on the m.loop.md_level? How should I judge what's
best for my loop?
Generally speaking, the longer the refinement you can afford, the
better. But for longer loops the limiting factor is almost certainly the
conformational sampling, which is probably easier to control by building
more models rather than longer refinement.