pychemcurv is a python package for structural analyzes of molecular systems or solid state materials focusing on the local curvature at an atomic scale. The local curvature is then used to compute the hybridization of molecular orbitals.

The main features of the library are available from a Plotly/Dash web application available here: The web-app allows to upload simple xyz files and compute the local geometrical properties and the hybridization properties. The application source code is available in a separate repository at pychemcurv-app.


Pyramidalization angle of a \(C_{28}\) fullerene mapped on the structure with a colorscale.


Pychemcurv is divided in two parts. The first one is a standard python package which provides two main classes to compute the local curvature at the atomic scale and the hybridization of a given atom. Second, a Plotly/Dash web application is provided in order to perform a geometrical and electronic analyzes on molecules or materials.

The web application is available at The web-app allows to upload simple xyz files and compute the local geometrical properties and the hybridization properties. The application source code is available in a separate repository at pychemcurv-app.

Some jupyter notebooks are provided in the notebooks/ folder and present use cases of the classes implemented in this package. You can access to these notebooks online with binder.

binder notebooks

Citing pychemcurv

Please, consider to cite the following paper when using either the pychemcurv library or the web application.

Julia Sabalot-Cuzzubbo, Germain Salvato Vallverdu, Didier Bégué and Jacky Cresson Relating the molecular topology and local geometry: Haddon’s pyramidalization angle and the Gaussian curvature, J. Chem. Phys. 152, 244310 (2020).



Before installing pychemcurv it is recommanded to create a virtual environment using conda or virtuelenv.

Short installation

Using pip directly from github, run

pip install git+git://

Alternatively, you can first clone the pychemcurv repository

git clone

and then install the module and its dependencies using

pip install .

Full installation

If you want to use the web application locally or if you want to use nglview to display structures in jupyter notebooks you need to install more dependencies. The setup configuration provides the viz and app extras so, using pip, run one of

pip install .[app]
# or
pip install .[viz]
# or all extras
pip install .[app, viz]

# escape square bracket with zsh
pip install .\[app, viz\]

If you have installed nglview you have to enable the jupyter extension

jupyter-nbextension enable nglview --py --sys-prefix

The files requirements.txt and environment.yml are provided to setup a full environment with all dependencies. Using pip, in a new environment you can run the following command to install dependencies

pip install -r requirements.txt

or using conda you can create the new environment and install all dependencies in one shot by

conda env create -f environment.yml

The name of the new environment is curv.

Do not forget to enable the jupyter nglview extension (see above).

Install in developper mode

In order to install in developper mode, first create an environment (using one of the provided file for example) and then install using pip

pip install -e .[app, viz]

If you want to build the documentation you also need to install sphinx.

Run the web application

The web application is available in this separate repository: pychemcurv-app The main aim of the application is to use the pychemcurv package and visualize the geometrical or chemical atomic quantities mapped on the chemical structure of your system.

The application is available online at this address:

Demo video:

In order to run the application locally, you have to install all the dependencies and in particular dash and dash-bio. You can do that from the files requirements.txt or environment.yml or directly using pip.

Then, clone the github repository on your computer

git clone

To run the application, change to the pychemcurv-app/ directory and run the file.

[user@computer] (curv) > $ cd pychemcurv-app/
[user@computer] (curv) > $ python
Running on
Debugger PIN: 065-022-191
* Serving Flask app "app" (lazy loading)
* Environment: production
WARNING: This is a development server. Do not use it in a production deployment.
Use a production WSGI server instead.
* Debug mode: on

Open the provided url to use the application.

You can switch off the debug mode by setting debug=False on the last line of the file.

Common error on local execution

If the application does not start with an error such as:

socket.gaierror: [Errno 8] nodename nor servname provided, or not known

Go to the last lines of the file and comment/uncomment the last lines to get something that reads

if __name__ == '__main__':
    app.run_server(debug=True, host='')
    # app.run_server(debug=False)