ADL: Long Simulation Time (~180 hrs)

Van Dam, Hubertus hvandam at
Wed Oct 28 08:25:18 PDT 2020

Hi Joey,

I don't think splitting the grid is even possible. Every ligand needs to be docked against the potential of the whole grid. The more obvious question is: How many ligands are you trying to dock. Docking two different ligands is completely independent. So you can split your ligand set into multiple smaller sets and dock all sets concurrently (assuming you have a parallel cluster).

Best wishes,


Hubertus van Dam, 631-344-6020, hvandam at
Brookhaven National Laboratory

On 10/27/20, 23:07, "autodock-bounces at on behalf of Joey Farrell" <autodock-bounces at on behalf of jpfarrell at> wrote:


    I am running a flexible docking job with files generated by AutoDockTools using Autodock Vina as the simulation-driving software.

    The grid space size is 40 x 40 x 48, and I believe I have already exceeded the efficient ratio of CPU to exhaustivity. Are there any natural suggestions to make this job more feasibly runnable (i.e. ~ 36 hours runtime)? Currently, it seems projected to run for about 180 hours. I suppose the most natural idea might be to subdivide this grid into smaller grid simulations; if this is the case, is there any recommended software to automatically splice this volume into N disjoint subgrids for use with Vina?

    Thank you in advance.

    --- ADL: AutoDock List  ---;!!P4SdNyxKAPE!RJ6fdzIO-WYQmZbbVwF2pz6Xn0zDcmjEHqIQvbtuKbeKjM0nTL5drRpPB32n4ac$  ---

More information about the autodock mailing list