Getting started

LobsterPy is a package that enables automatic analysis of LOBSTER outputs to get summarized bonding information and relevant bond plots. Additionally, one can also generate features for machine learning studies from LOBSTER outputs. One can download LOBSTER from http://www.cohp.de.

Please note that LobsterPy relies on the LOBSTER computation output files. Thus, it will be only able to analyze data that has been computed in the LOBSTER run.

Installation

Python version

Before the installation, please make sure that you are using one of the supported Python versions (see pyproject.toml).

Standard installation

Install using pip install lobsterpy

Installation with featurizer

Install using pip install lobsterpy[featurizer]

Contributing guidelines / Developers installation

A short guide to contributing to LobsterPy can be found here. Additional information for developers can be found here.

Basic usage

• Automatic analysis and plotting of COHPs / COBIS / COOPs:

  

You can use lobsterpy description for an automated analysis of COHPs for relevant cation-anion bonds or lobsterpy automatic-plot to plot the results automatically. It will evaluate all COHPs with ICOHP values down to 10% of the strongest ICOHP. You can enforce an analysis of all bonds by using lobsterpy automatic-plot --allbonds. You can also switch the automatic analysis to use the ICOBIs or ICOOPs. You need to add --cobis or --coops along with the mentioned commands for e.g.like lobsterpy description --cobis

An interactive plotter is available via lobsterpy automatic-plot-ia.

Currently, the computed Mulliken charges will be used to determine cations and anions. If no CHARGE.lobster is available, the algorithm will fall back to the BondValence analysis from pymatgen.

Please be aware that LobsterPy can only analyze bonds that have been included in the initial Lobster computation. Thus, please use the cohpgenerator within Lobster (i.e., put cohpGenerator from 0.1 to 5.0 in the *lobsterin).*

It is also possible to start this automatic analysis from a Python script. See "examples" for scripts.

• Plotting DOS from LOBSTER computations:

To plot densities of states obtained from LOBSTER use lobsterpy plot-dos.

• Generic COHP/ COOP / COBI plotter:

We included options to plot COHPs/COBIs/COOPs from the command line. lobsterpy plot 1 2 will plot COHPs of the first and second bond from COHPCAR.lobster. It is possible to sum or integrate the COHPs as well (--summed, --integrated). You can switch to COBIs or COOPs by using --cobis or --coops, respectively.

• Other command line tools:

  lobsterpy create-inputs will create standard inputs based on existing POSCAR, POTCAR, and INCAR files. It will allow testing for different basis sets that are available in Lobster. This feature is currently only available for PBE_54 POTCARs, as only the pbeVASPfit2015 basis in LOBSTER that has been fitted to PBE POTCARs includes additional orbitals relevant to solid-state materials. Please check out our publication https://doi.org/10.1002/cplu.202200123 and LOBSTER program manual for more information

• Further help?

  You can get further information by using lobsterpy --help and also by typing lobsterpy description --help, lobsterpy automatic-plot --help, lobsterpy plot --help.

Documentation

• Checkout the documentation and tutorials for more details.

How to cite?

Please cite our paper: A. A. Naik, K. Ueltzen, C. Ertural, A. J. Jackson, J. George, Journal of Open Source Software 2024, 9, 6286. https://joss.theoj.org/papers/10.21105/joss.06286. Please cite pymatgen, Lobster, and ChemEnv correctly as well.

You can find more information on the methodology of the automatic analysis in J. George, G. Petretto, A. Naik, M. Esters, A. J. Jackson, R. Nelson, R. Dronskowski, G.-M. Rignanese, G. Hautier, ChemPlusChem 2022, 87, e202200123. https://doi.org/10.1002/cplu.202200123.

LobsterPy is now a part of an atomate2 workflow

We have now also included the automatic analysis into a fully automatic workflow using VASP and Lobster in atomate2. More documentation and information will follow soon.

Acknowledgements

The development of the program has been supported by a computing time grant. We gratefully acknowledge the Gauss Centre for Supercomputing e.V. (www.gauss-centre.eu) for funding this project by providing computing time on the GCS Supercomputer SuperMUC-NG at Leibniz Supercomputing Centre (www.lrz.de) (project pn73da).