search

ccsb-scripps / AutoDock-Vina

AutoDock Vina
http://vina.scripps.edu
Apache License 2.0
559 stars 199 forks source link

Method to capture receptor structure state after flexible docking #319

Open eunos-1128 opened 3 weeks ago

eunos-1128 commented 3 weeks ago

Hi,

I'm interested in capturing the exact state of the receptor after flexible docking.

It seems there is no description about that on docs, but is there any way to do that with vina?

If not, does any other methods possible exist?

Regards,

rwxayheee commented 3 weeks ago

Hi @eunos-1128 What are some specific metrics you would be interested in using? Sidechain RMSD? Dihedral angles (and clustering by rotamers)? Many would use these to analyze MD trajectories of proteins. If you are able to export the after-docking structures, programs like pytraj, ProDy, or MDAnalysis have built-in functions for these measurements. Otherwise, you can also compute them by yourself using the output coordinates. (I personally think it's a very good idea to regroup results by distinct receptor conformations. I just don't know a very straightforward way to go from AutoDock right now

eunos-1128 commented 3 weeks ago

@rwxayheee

What are some specific metrics you would be interested in using? Sidechain RMSD? Dihedral angles (and clustering by rotamers)?

Both of them.

Besides calculating them, I want to take closer look at differences among conformers using molecule viewer in detail.

diogomart commented 3 weeks ago

With mk_export.py the sidechains are written as additional molecules to the output SDF file (they are together with the ligand).

Meeko v0.6 will be able to write a PDB of the entire protein with the updated positions of the sidechain. This is better for visualization and for integrating with other software, such as starting an MD simulation from docked poses.

@eunos-1128 is this the feature/information you were looking for?

eunos-1128 commented 3 weeks ago

With mk_export.py the sidechains are written as additional molecules to the output SDF file (they are together with the ligand).

Meeko v0.6 will be able to write a PDB of the entire protein with the updated positions of the sidechain. This is better for visualization and for integrating with other software, such as starting an MD simulation from docked poses.

@eunos-1128 is this the feature/information you were looking for?

@diogomart Yes, this is what I was looking for.

I'll try the script of meeko :)

eunos-1128 commented 2 weeks ago

@diogomart

I could confirm sidechains in SDFs using mk_export.py Thank you.

Meeko v0.6 will be able to write a PDB of the entire protein with the updated positions of the sidechain. This is better for visualization and for integrating with other software, such as starting an MD simulation from docked poses.

I want the structure of the entire protein right now.

In Meeko's repo, there is a tag, called v0.6.0-alpha.3. Has it had that feature already?

I looked into the code but not sure of that.

diogomart commented 2 weeks ago

The alpha versions do not have that feature. Only v0.6 (non-alpha) will. The alpha were created to make some receptor preparation features more easily available before the a stable v0.6 is released.

eunos-1128 commented 2 weeks ago

@diogomart

OK, Thank you. Until v0.6 is released, I'll manage with current version's meeko.