Releases: molssi-seamm/forcefield_step
Bugfix: error submitting jobs with local forcefield files
Bugfix: factor of 2 for dihedrals and impropers in ligpargen
- The ligpargen tool was missing a factor of 2 in the dihedral and improper parameters.
- Corrected the search paths for forcefields.
- Improved the documentation.
Added customizable local forcefields and access to LigParGen
- Added the machinery to handle local forcefield files in either ~/.seamm.d/data/Forcefields (personal) or ~/SEAMM/data/Forcefields (site).
- Added 'ligpargen' command to access custom parameters from the LigParGen service at Yale University, ading them to the 'ligpargen.frc' personal forcefield, which is automatically included in 'oplsaa.frc' if it exists.
Bugfix: PF6- angle parameters
The angle parameters for PF6- in the CL&P forcefield only work if the 180º F-P-F angles are not included in the calculation. Replacing them with an equivalent periodic SHAPES-like potential almost works; however, since 0º is a valid angle and there are no 1-3 nonbonds, nothing keeps the F atoms apart. This is solved using a tabulated potential based on the SHAPES potential but with an added 1-3 repulsion
large enough that the gradient is always pushing small angles apart, but not large enough to affect the minimum at 90º.
Fixed bugs and add imidazolium cations
- Fixed significant problems in yesterday's release, mainly the wrong units for torsions.
- Added CL&P parameters for 1-alkyl-3-methyl imidazolium cations
Added parameters for TFSI to CL&P/OPLSAA
- Parameters for TFSI - bis[(trifluoromethyl)sulfonyl]imide
Added more atom typing for OPLS-AA
Added typing for the following functional groups, which were missed in the first implementation.
- cyclopropane -CH2-, -CHR-, and -CR2-
- hexafluorobenzene
- difluorobenzene
- bromobenzene
- iodobenzene
- thiophenol
- alkyl nitriles
- nitroalkanes
- nitrobenzene
- methylene in phenylacetonitrile
- corrections to methylene nitrile anion
Added typing for fluorobenzene to OPLS-AA
The typing for this and few other atom types were forgotten in the first round. More will be added soon for e.g. perfluorobenzene
Fixed F-P-F angle for PF6- in OPLSAA+
The Lennard-Jones repulsion added to the angle was not quite sufficient to stop it having a metastable structure with an FPF angle of ~40º.
Fixed issues with angles in octahedral systems
- Implemented tabulated angle to fix the problem of the angle at 0º in the simple fourier angle. Added a LJ repulsive 1/R^12 term between the terminal angles of the angle.