Under active development - do not use
Semi-automatic identification of pharmacophore features from unbound receptor structure using gnina fragment docking.
Currently installation should be done by cloning this GitHub repository as well as following the installation instructions for GNINA 1.0 and rd-kit.
To install simply do the following command:
git clone https://github.com/gnina/phinder.git
Once you have cloned the repository you can work with the phinder scripts to generate pharmacophore!
Currently, Phinder is meant to be run within a project directory of which it's structure is very important for easily working with various files generated from the program. The best way to run phinder is to do the following:
- Create a project and copy Phinder into said project
mkdir projectname
cp -r Phinder projectname/Phinder
- Go into said project
cd projectname
Docking is done using GNINA 1.0 where we have locked in a set of probes for the fragment docking. Those probes are as follows:
- Acetate
- Acetamide
- Benzene
- Guanidine
- Imidazole
- Isobutane
- Isopropanol
- Isopropylamine
- Water
Docking is performed if no fragment docked files are detected (fragmentDocked.sdf).
To phind pharmacophores use the Phinder.py file on your receptor of choice. All you have to do is specify the locations of the receptor, ligand, and output directory. Phinder.py will automatically dock and generate all pharmacophore features in a single file (GeneratedPharma.json) in the output directory that was specified. To run phinder use the following command:
## From Phinder Home Directory
python Phinder.py -r receptor -l ligand -o output_directory
You will see the GeneratedPharma.json file in the output directory!
Congrats! You have completed your first Phind! Now you can review your Pharmacophores in PyMOL. To do so make sure you have the load_query module for PyMOL it can be found here