[Dock-fans] Fwd: grid size (former "parameters" and "very large receptors")
Trent E. Balius
tbalius at aol.com
Sat Jan 22 11:50:31 PST 2011
I have an incomplete understanding of the orient code. I believe the way that the anchor placement and sphere matching works is that the dock matches say 3 atoms of your ligand with 3 spheres and minimizes the residuals. The code is fairly complicated and there is clustering and pruning going on. Perhaps others can common on this.
The fact that you are getting a segmentation fault and not just could not complete growth is worrisome. If you can send me your files I can see if I can replicate this error.
I have observed a case were the anchor was placed with 3 spheres and was not well overlapped with any because of the sparse spheres. however this is not common, usually one atom is placed on top of one sphere. My earlier comment might occur if you had one sphere per pocket.
The code should be able to perform blind docking. In fact it maybe better than audodock since dock performs incremental growth while audodock uses a genetic algorithm. however this claim is untested.
Trent E. Balius
Graduate Student, Rizzo Group,
Department of Applied Mathematics and Statistics,
Stony Brook University.
Office: Math Tower 3-129, Phone: (631) 632-8519
From: Francesco Pietra <chiendarret at gmail.com>
To: Trent E. Balius <tbalius at aol.com>
Cc: dock-fans <dock-fans at docking.org>
Sent: Sat, Jan 22, 2011 9:34 am
Subject: Re: [Dock-fans] Fwd: grid size (former "parameters" and "very large receptors")
Thanks a lot for the time you spent answering me. That we examine
everything graphically, sphere included, is our standard routine, in
particular the various sphere clusters I alluded to. My post was to
show how things run (or not run) and the very reason for posting was
the last sentence:
""Finally, why "When you have spheres in multiple pockets dock may try
and orient the ligand to spheres located in different pockets causing
the ligand to be placed in a vary unfavorable position and pruned
leaving you with no orients"? Is that still inadequacy of the code
(something that could be fixed) or that is a more serious issue?""
which perhaps escaped attention. What I would aim to know is whether
these are limitations (I have just transcribed limitations that you
pointed out before, using just your words) by the code, limitations
that could be fixed, or what I would like to do is beyond the
capability of the code.
I have now checked the autodck4 has no such limitations, although its
use of partial charges and grid space pose other problems.
On Sat, Jan 22, 2011 at 2:30 PM, Trent E. Balius <tbalius at aol.com> wrote:
> See manual for discussion of the spheres:
> Here is an excerpt from the SphgenOverview section:
> "These sets, or clusters, are sorted according to numbers of constituent
> spheres, and written out in order of descending size. The largest cluster is
> typically the ligand binding site of the receptor molecule. The program
> showsphere writes out sphere center coordinates in PDB format and may be
> helpful for visualizing the clusters (see showsphere)."
> Here is an excerpt from the discussion in the SphgenOutput section of the
> The clusters are listed in numerical order from largest cluster found to the
> smallest. At the end of the clusters is cluster number 0. This is not an
> actual sphere cluster, but a list of allof the spheres generated whose radii
> were larger than the minimum radius, before the filtering heuristics ( i.e.,
> allowing only one sphere per atom and using a maximum radius cutoff) and
> clustering were performed. Cluster 0 may be useful as a starting point for
> users who want to explore a wider range of possible clusters than is
> provided by the standard sphgen clustering routine.
> When things start not working it is always a good idea to visualize
> everything with a molecular viewing program like Chimera.
> Cluster 0 should cover the whole receptor. Other clusters should be specific
> to a pocket as defined by the spacial clustering of the spheres.
> I strongly encourage you to visualize the clusters on your protein with
> chimera or your favorite program to see the differences. Show the grid box
> as well.
> Segmentation faults are always extremely frustrating.
> I am not an expert on amberscore, but I am not sure why you can't compare
> things docked to multipule sphere sets as long as you use the same
> Good luck and I hope this helps,
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Dock-fans