forked from fzahari/ParFit
-
Notifications
You must be signed in to change notification settings - Fork 0
ParFit automates the process of fitting molecular-mechanics parameters to data obtained by ab-initio calculations.
License
drmperez/ParFit
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Development of ParFit has been supported in part by the Critical Materials Institute, an Energy Innovation Hub funded by the U.S Department of Energy (DOE), Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, in part by the Chemical Sciences, Office of Basic Energy Sciences, Office of Science, DOE, and in part by the Laboratory Directed Research and Development Program, Fundamental Science Division of the Pacific Northwest National Laboratory. This file contains the basic information to get started. Refer to Doc/Manual.pdf for more detailed information. The main directory, ParFit, contains subdirectories: Data, Doc, ParFit, and Utils. The Data subdirectory contains the Engine, Gamess, and ParFit subdirectories. The Data/Engine directory contains the mmengine input, output, and parameter files. The Data/Gamess directory contains the QM molecular information. The Data/ParFit subdirectory contains the energy profile and parameters at optimization intervals. The Doc subdirectory contains the ParFit user manual, Manual.pdf. The ParFit subdirectory contains the Python scripts and example input files with root filename "dih_scan_inp." ParFit should be run from this directory using the command: ./Parfit.py The Utils subdirectory contains an interactive ParFit input file generating program (PFinp.py) and template scripts that can rename multiple files (rename.py), automatically submit multiple jobs on a typical cluster (submit.py) and visualize the energy profile fit using matplotlib (QM_vs_MM_energies.py). ParFit.py is the main program in the ParFit subdirectory. ParFit uses an input file named dih_scan_inp. One can generate an input file using the PFinp.py program. The input file format is as follows: Line/Block Description --------------- -------------------------------------------------------------- 1 optional |Two values, "mult", and the number of molecules for which to |fit. Example: |mult, Y |where Y is the the number of molecules to use for parameter |fitting, Y must be 1 or greater. | 2 to N |the full/comp file type, the root filename, the atoms forming |the dihedral angle, and the triplet defining the dihedral |angle. | |"full/comp QM data”: if "full" all the Gamess output files |must be in the Data/Gamess subdirectory. If "comp,” a compact |form of the QM data should be contained in one file in the |/Data/Gamess subdirectory. | |"the root filename": the name without "X.log" where "X" |is the dihedral angle, if "full,” or the name without "scan,” |if "comp.” | |"the atoms forming the dihedral angle": four numbers defining |the dihedral angle | |"a triple defining the dihedral angle range": initial, final, |and angle step size. | N + 1 |The full mmengine path. | N + 2 |MM-type: "mm3" or "mmf94" | N + 3 |Optimization algorithm: "ga," "fmin," or "hybr" | N + 4 to M |The choice of the parameters to be optimized according to the |following format: |line number in add_MM3.prm or add_MMFF94.prm, 'c' or 'p', i.e. |"constant parameter" and "parameter to be optimized," |respectively. |i.e, "15 c c p" means that within line 15, the first and second |parameters will remain constant and only the third parameter |will be optimized. | M + 1 last line |Include "csv_on" to request comma separated value snapshot |files to be printed in the /Data/ParFit subdirectory. Each |snapshot file contains the relative abinitio and MM energy for |each dihedral angle. If the file is not needed, use "csv_off". Sample inputs can be found in the ParFit subdirectory. The "dih_scan_inp_*_mm3" are sample inputs for MM3 and "dih_scan_inp_*_mmf94" for MMFF94. To use templates, copy to a different filename and change the mmengine path.
About
ParFit automates the process of fitting molecular-mechanics parameters to data obtained by ab-initio calculations.
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published
Languages
- Python 100.0%