repESP
?repESP
is a program used in computational chemistry to investigate different methods of assigning partial charges.
It takes the name from reproducing the (molecular) electrostatic potential (ESP) from partial charges.
Partial charges can be extracted from the output of Gaussian 09 or provided by the user.
Current capabilities of the program include:
cube
files)The program was developed as part of Jan Szopinski's MSci project at the Hunt Research Group, Imperial College London. The program was used to investigate and develop the following novel concepts:
repESP
repESP
implemented a lot of functionalities, which can be used to examine the various methods of assigning partial charges.
Many of these functionalities are exposed to users as command line (CLI) scripts and are easy to use.
The scripts perform well-defined, modular tasks and can be woven together to create one's own work flow (Unix style).
The following chart presents some of the repESP
capabilities:
The monospaced text above arrows shows the names of the relevant command line scripts.
When the out-of-the-box functionalities are not sufficient or flexible enough, users can create their own scripts by using the repESP library.
The overview of the available scripts can be found here and the command line interface help can be found here. Library documentation is hosted at http://jszopi.github.io/repESP.
repESP
The following instructions are for Unix-like systems, like MacOS and Linux.
There shouldn't be anything stopping repESP
from running on Windows but that hasn't been tested.
repESP
needs input from computational chemistry software to obtain the ESP and partial charges.
Currently only the output from Gaussian 09 is known to be recognized, please open issues for other Gaussian versions or other software.
Further, to run repESP
you need Python 3.7 and two other free programs, which need to be in your PATH
.
resp
resp
is a free program necessary to perform ESP-fitting.
Download link can be found at the bottom of this documentation page.
To install you'll have to compile the Fortran source and put the resulting binary in your PATH
.
resp
and repESP
need information about atom equivalence in the form of .respin
files.
AmberTools is a suite of free programs (link).
However, it requires compilation from C, C++ and Fortran sources, which proved tricky on MacOS.
To create the necessary input for a methane molecule, first create the .ac
file from Gaussian .log
using antechamber
:
antechamber -i methane.log -fi gout -o methane.ac -fo ac
Then run respgen
twice:
respgen -i methane.ac -o methane.respin1 -f resp1
respgen -i methane.ac -o methane.respin2 -f resp2
Installing repESP
is standard and simple:
~/bin
)git clone https://github.com/jszopi/repESP
pip3 install .
With the following caveats regarding the final point:
master
branch.
Passing -e
to pip3 install
will ensure that the repESP
version installed reflects what's in your local git repo.
To later update the code to latest version run: git fetch --all ; git pull
.[docs]
to the install command, i.e. run pip3 install .[docs]
instead.repESP
is free software released under the Gnu Public Licence version 3.
All contributions are very welcome, from improving the content of this overview to tweaking the code.
There's still plenty you can do! First, have a look at GitHub's "How to contribute".
git
git
This program needs some work to get where it wants to be. If you want to help out, please coordinate with the maintainer.