Required inputs

PDB files: these should be aligned on the target ligands. Two examples are provided on GitHub, one for protein-protein interactions and one for small molecule-protein interactions. If you choose to upload your own PDB files, you will first need to clean, standardize, and align input PDB files for seamless usage in MAGPIE (See the Preparing your own dataset for use in MAGPIE and Helper Scripts sections of this documentation).

Target chain ID from PDB: the program uses the first file in the directory to upload the target ligand structure.

Protein binder chain ID from PDB: this chain identifier must be the same across all PDB files.

Target type: indicate whether the target ligand is a small molecule or a protein.

Target residue index (for protein ligands) or unique atom names (for small molecule ligands): provide the target residue indices for proteins or unique atom names for small molecules. Alternatively, you can input ‘all’ to consider all AAs/heavy atoms.