Generating FODs: FODLego

FODLego is a deterministic method to generate FODs for a given molecular structure. It uses monoatomic radii and a series of heuristics to predict FODs and it employs a class-based paradigm that allows the user to modify the construction parameters used to define FOD points around a molecule. It only generates FODs up to \(\text{3sp}^3\) orbitals.

Installation and Running

FODLego is available to download via pip :

pip install FODLego

FODLego should now be available as command on your shell. If you execute fodlego with no input, you should get the following error:

WARNING:__main__.py:main(): No arguments were given, please provide XYZ file name

For CLUSTER files, you simply call:

fodlego CLUSTER

For an .xyz file, you simply need to input the filename:

fodlego input.xyz

SMILES structures also work. You just need to add quotation marks around it.

fodlego  "NC(=S)NC"

For SMILES structures, you can check RDKit’s documentation to read more about the rationale of 3D structure approximation.

Sample Performance on Small Organic Molecules

The following image is taken from Angel Villegas’ thesis A Cheminformatics-Based Approach to the Prediction of Fermi-Orbital Descriptors (FODs) in Fermi-Lowdin Orbital Self Interaction Correction (FLOSIC) Theory

Development Reference

Github

Please visit the Github page in order to suggest features, improvements, or ask questions!

Class Design