Supporting Information for "Electrostatic Embedding of Machine Learning Potentials".

For practical implementation of EMLE method see emle-engine package.

All the required files can be found at https://doi.org/10.5281/zenodo.7051785.

train.ipynb

Training protocol. Requires:

• qm7.mat
• QM7_B3LYP_cc-pVTZ.tgz
• QM7_B3LYP_cc-pVTZ_horton.tgz.

mpro.ipynb

Emedding energy prediction for SARS-CoV-2 MPro complex with PF-00835231. Requires:

• mlmm.mat
• mpro_xyz.tgz
• mpro_B3LYP_VTZ.tgz
• mpro_PBE0_VTZ.tgz
• mpro_B3LYP_VDZ.tgz
• mpro_BLYP_631Gs.tgz
• mpro_XTB.tgz
• mpro_AM1.tgz
• mpro_PM3.tgz