This code attempts to predict the influence of osmolytes on protein stability, using the methods in: Auton M, Bolen DW. Predicting the energetics of osmolyte-induced protein folding/unfolding. Proc Natl Acad Sci 102:15065 (2005) https://doi.org/10.1073/pnas.0507053102
Other models may be added in the future.
Regardless of how you install the software, we recommend doing so in a virtual environment. You can create a virtual environment using, e.g., one of the following:
conda create -n osmolytes pip
(note that you will need the pip package for later steps) or
virtualenv -p python3 ~/venvs/osmolytes
You will need to activate the virtual environment prior to installing or using software, e.g., with one of the following:
conda activate osmolytes
or
. ~/venvs/osmolytes/bin/activate
Most users will want to install the software from the PyPI repository via the following steps:
- Activate your virtual environment as described above
- Install the package using
pip install osmolytes
Adventurous users and developers may want to install the software from source code via the following steps:
- Activate your virtual environment as described above
- Download the source code
git clone https://github.com/Electrostatics/osmolytes.git
- Change to the source code directory (e.g.,
cd osmolytes) - Install the package from source code using
pip install -e .
If you are using conda, you may first need to install pip with
conda install pip
Installing from source also allows testing of the software with
python -m pytest .
These tests check the integrity of the software and provide information about the accuracy of the various calculations (particularly, surface area).
The main command-line tool installed by this package is the program mvalue.
The input to the mvalue program is a PQR file which can be generated by the PDB2PQR software.
Most users will want to use the PDB2PQR web server.
The mvalue program has been tested with PQR files generated by PDB2PQR using the following settings:
- Protonation states assigned by PROPKA
- PARSE force field
- Internal naming scheme
In addition to generating a PQR, the PROPKA software generates additional output in the form of *.propka files that contain useful predictions about the pH stability of the protein.
The mvalue program has several options which can be listed using
mvalue --help
The basic usage of mvalue is
mvalue 2BU4.pqr
where 2BU4.pqr is the name of an example PQR file.
The mvalue program produces output that describes
- The residue composition of the protein
- The contribution of each residue type (and its surface area) to the m-value of the protein for each osmolyte species (at 1M concentration).
- The m-value associated with each osmolyte at 1M concentration.
The concept of a m-value is described in Wikipedia and many review articles on protein folding and stability.
Re-implementation of these algorithms was supported by National Institutes of Health grant GM069702.