.. include:: ../../toolkit-substitutions
.. _section_szybki_releasenotes_173:

Szybki TK 1.7.3
================================================================================

New features
--------------------------------------------------------------------------------
* A new forcefield for protein-ligand interactions has been added. It is a combination
  of MMFF94S (or MMFF94) with Amber. In this combined force field, MMFF94S is used 
  to describe the intramolecular interactions of the ligand and the Amber force
  field is used for the VdW and Coulomb interactions between ligand and protein.
  Currently, this force field can be used only for ligands inside rigid proteins.

* Two new methods were added to the API: :oe:meth:`OESz::OESzybki::SetTorsionConstraint` 
  and :oe:meth:`OESz::OESzybki::ClearTorsionConstraint`.  The first one adds a
  single-minimum harmonic torsional constraint for a torsion which is defined by
  a SMARTS pattern. It uses the functional form:
  :math:`V=k_c(cos(phi) - cos(phi0))^2`, where :math:`k_c` is the user specified force
  constant and :math:`phi0` is the reference torsion dihedral angle. The second method
  removes the torsional constraint if it exists.

* Estimation of ligand entropy using analytical MMFF Hessian is extended for 
  ligands bound in a rigid protein. Previously the entropy of only free ligands in
  vacuum or in solution could be estimated with analytical second derivatives. 

Major bug fixes
--------------------------------------------------------------------------------
* A few numerical errors in the calculation of analytical second derivatives
  (Hessian matrix) in MMFF torsion and out-of plane terms have been fixed.


Minor bug fixes
--------------------------------------------------------------------------------
* Constraint energy was reported twice by the method :oe:meth:`OESz::OESzybkiResults::Print`.
  The problem is fixed.

* The method ``OESz::OESzybki::GetExactVdWProteinLigand`` was misspelled
  (``GetExactVdEProteinLigand`` instead of ``GetExactVdWProteinLigand``). 

* When preoptimized ligands are used for entropy estimation it is necessary to gently 
  move the ligand out of the minimum. The gradient normal below which such a perturbation was 
  done was raised from 1 kcal/(mol |A|:math:`\AA`) to 10 kcal/(mol |A|:math:`\AA`). 
  Research showed that for some conformations the above limit was too low which resulted 
  in overestimation of its vibrational entropy.
