search

choderalab / yank

An open, extensible Python framework for GPU-accelerated alchemical free energy calculations.
http://getyank.org
MIT License
181 stars 71 forks source link

Binding site identification? #1182

Open jadeshi opened 5 years ago

jadeshi commented 5 years ago

Hi, possibly a very dumb question here, but I was running through the T4-lysozyme / p-xylene example, and I noticed that when I load the receptor PDB and ligand mol2 simultaneously in PyMol, the ligand doesn't seem to be bound to any site and is fairly distant from the protein. Also in the .yaml file that there doesn't seem to be any place to specify the coordinates of the L99A binding site. So I was wondering, how does YANK know what the correct binding site is, or is it sampling all potential sites and generating an ensemble average?

Thanks!

andrrizzi commented 5 years ago

Hi @jadeshi . The calculation will start from the coordinates of the files. YANK doesn't try to guess the binding site, but that example will create a harmonic restraint between the centroid of the receptor and ligand that should bias the ligand in the direction of the receptor.

In practice, for regular calculations, the recommended way is to start from a docked/crystal pose, which should be given in the mol2 file.

You implicitly define what is the binding site by setting the restrained parameters. In particular, you can set the restrained_receptor_atoms option in the harmonic restraint (see the docs: http://getyank.org/latest/yamlpages/experiments.html).

smutao commented 1 year ago

PDB id: 225L could be a good reference