Batch docking box size

[alanhala]

Hi!

I want to run a batch docking of N ligands over one protein in one pocket or cavity. In order to achieve that I calculate the bounding box of that cavity and then I move manually the ligand to the center of that bounding box. After that I calculate the new bounding box in which both ligand and pocket fit.

If I start using this option, I might need different bounding box becuase ligands might be bigger and might not all fit in the same bounding box. That makes me think if my original idea was ok, meaning that if it is needed that the ligand is fully contained inside the bounding box, or the bounding box just needs to be the area of the protein where I want to perform the docking.

If the ligand has to be inside the bounding box, then what's the recommendation for using this batch option?

Thanks in advance!

[diogomart]

Hello,

No need to move the ligand inside the box, docking will take care of that. The box just needs to be the area of the protein where you want to dock, and it needs to be large enough to accommodate the ligands because there is a very large penalty for atoms that are outside the box.

[alanhala]

Hi Diego,

Thanks for you quick response!

No need to move the ligand inside the box, docking will take care of that.

Interesting. I though the ligand should have been in the box because of what you mentioned later (I read about the penalty somewhere). So in my pipeline I was moving the ligands to the center of the bounding box.

and it needs to be large enough to accommodate the ligands

Regarding this... It might be a silly question, but when it comes to performing docking against one pocket trying with different ligands, how do you make sure that each of them can fit within the box? Because I assume that you don't want a really big box because that might have impact on the result, right?

[diogomart]

Docking will move the ligand around and it will be completely inside the box as long as it fits. Typically we dock molecules around 15 to 30 non hydrogen atoms, more or less, and boxes tend to be around 15 to 30 Angstrom in each dimension. Can't be 100% sure every ligand fits, or that better binding modes wouldn't arise with a larger box, but we visually inspect docked poses and if we think some ligands would have benefited from a larger box then we make it larger.

[alanhala]

Right, so basically after each docking you could check if the pose is still inside the box. If it is not then I can assume that the pose got some penalty and maybe I should retry with a bigger box?