Port RF Boresch restraints to PME

[nigel-palmer-cresset]

Based of Boresch restraints for RF in
src/libs/SireMove/openmmfrenergyst.cpp and
43ed4bc

• I confirm that I have merged the latest version of devel into this branch before issuing this pull request (e.g. by running git pull origin devel): y
• I confirm that I have added a changelog entry to the changelog (we will add a link to this PR as part of the review): y
• I confirm that I have permission to release this code under the GPL3 license: y

Suggested reviewers:

@chryswoods, @lohedges

[lohedges]

Thanks, @nigel-palmer-cresset. This looks good to me. Happy to wait for @fjclark to comment before merging if needed. (He has also ported this on his own branch.)

[fjclark]

Hi @lohedges the Boresch part looks good to me. My only minor comment is that the flat-bottomed distance restraint equation:

    OpenMM::CustomBondForce * custom_link_bond = new OpenMM::CustomBondForce("lamrest*kl*max(0,d-dl*dl);"
                                                                             "d=(r-reql)*(r-reql)");

is inconsistent with that in openmmfrenergyst.cpp:

    OpenMM::CustomBondForce *custom_link_bond = new OpenMM::CustomBondForce("delta(min(0, r_eff))*(lamrest^5)*kl*r_eff^2;"
                                                                            "r_eff=abs(r-reql)-dl");

Likely because I updated the former to the latter and the older version has been copied. The minor issue with the older version is that the potential is not harmonic with equilibrium position dl (instead it's (r-reql)^2 - dl^2). I'd suggest updating to the newer version which gives the intended harmonic potential ((abs(r-reql)-dl)**2). Scaling with lambda^5 makes makes turning on the potential smoother.

[nigel-palmer-cresset]

Thanks @fjclark, I have updated the equation.