========================================================
- Go to: https://zhanggroup.org/GLASS/, and download the interactions_active.tsv, ligands.tsv, and targets.tsv files.
- Save the files to a folder called: TSV_2; after having done so, they'll be ready to merge.
** The following python libraries are required to run the model **
- Python 3.13
- Deepchem 1.x (requires RDKit)
- Scikitlearn (For MLPClassifier)
- Seaborn (For jointplot of XlogP and Molecular Weight)
- TSNE (For visualization of high-dimensional data)
** Dopamine D3 Autodock structure example **
- Tutorial for Autodock Vina: https://vina.scripps.edu/tutorial/
- After you've watched the tutorial, input the SMILES from your ligand of choice into the Ligand_and_Docking.py file. After having done so, It'll then output a .pdb file; this file will act as the ligand in the Autodock simulation.
- After you've dealt with the ligand, you're ready to setup the Protein docking structure. The way in which you go about doing this is by inputting the pdb_id into the Ligand_and_Docking.py file; after having done so, It should save a file named in accordance with the pdb_id.
** Example usage video **