The Molecular Graphics Laboratory Forum

AutoDock, AutoLigand, MGLTools, Vina, PyRx and more.
It is currently Sun Apr 22, 2018 1:08 am

All times are UTC




Post new topic Reply to topic  [ 4 posts ] 
Author Message
PostPosted: Fri Oct 23, 2009 8:16 am 
Offline
Micromolar User
Micromolar User
User avatar

Joined: Fri Sep 18, 2009 9:49 am
Posts: 13
I exported the AD4-docked conformations from dlg file and re-caculated the rms between these conformations and the crystal strcuture of the ligand by pymol .

However, the RMSD values from AD4 (generated when docking) and pymol (generated by align/rms/rms_cur command) are different.

Does anybody know how the rmsd in AD4 is defined/calculated? and how to calculate rmsd in the same way by any tool other than AD4? Thanks.


Top
 Profile  
 
PostPosted: Mon Oct 26, 2009 9:11 pm 
Offline
Micromolar User
Micromolar User
User avatar

Joined: Thu Apr 16, 2009 7:31 pm
Posts: 48
I assume you are specifying a reference structure for the calculations with the rmsref keyword?

See the following for a brief discussion of the rmsd calculations.
http://autodock.scripps.edu/faqs-help/f ... ng-results

Autodock's default rmsd calculation takes symmetry into account so that, for example, a carboxylate flipped 180 degrees from a template structure has 0A rmsd. This procedure gives lower rmsd values than calculating rmsd between atoms of exactly the same identity, which is, IIRC, what pymol does. You can force autodock to do it the second way by using the rmsnosym keyword in your dpf.


Top
 Profile  
 
PostPosted: Tue Oct 27, 2009 8:40 am 
Offline
Micromolar User
Micromolar User
User avatar

Joined: Thu Apr 16, 2009 7:31 pm
Posts: 48
stonefonly,

I'm going to answer your PM here for the benefit of future generations :P

You can use autodock to do the rmsd calculations with, e.g. Vina's output coordinates.

This is kind of kludgey - maybe some one knows of an easier way?

1st, load the docked conformation's coordinates into ADT. Go to Docking-->Ligand-->Choose and select the docked ligand. Then, go to Docking-->Ligand-->Ligand Parameters. In the parameter window, check "yes" for "Specify Initial Dihedrals?" and then uncheck all of the boxes that have the word "Random" next to them (i.e. for Initial position, relative dihedral offset, and dihedrals).

Now, set up a .dpf like the following:

---------------------------------------------------------------
ligand_types C OA
fld whatever.maps.fld
map whatever.C.map
map whatever.OA.map
elecmap whatever.e.map
desolvmap whatever.d.map
move blah-docked.pdbqt
about 27.976 -1.335 15.659
rmsref blah-crystal.pdbqt
tran0 27.976 -1.335 15.659
quat0 1 0 0 0
dihe0 0 0
ls_search_freq 0
set_sw1
do_local_only 1
analysis
-------------------------------------------------

"about" should be the same as "tran0", which are the xyz coordinates from the "User specified initial position" field in ADT's ligand parameter window. "move" is your docked conformation and "rmsref" is your crystallographic reference structure. "quat0" is "1 0 0 0" and "dihe0" is a zero for each torsion in your ligand.

Alter "ligand_types" , "fld", "map" etc. to fit your ligand and receptor. I don't think you really need to specify all of the ligand types and maps unless you need scores in addition to rmsd values.

When you run this, autodock will start a local search from the specified conformation. However, since the local search probability is set to zero, this initial conformation is never changed during the "search".

Your .dlg should show the rmsd from the reference near the end of the file like so:

--------------------------------------
Sorry! Unable to perform cluster analysis, because not enough conformations were generated.


MODEL 1
USER Run = 1
USER Cluster Rank = 1
USER Number of conformations in this cluster = 1
USER
USER RMSD from reference structure = 0.373 A
USER
USER Estimated Free Energy of Binding = -8.96 kcal/mol [=(1)+(2)+(3)-(4)]
----------------------------------------


Hope this helps!


Top
 Profile  
 
PostPosted: Tue Oct 27, 2009 9:54 am 
Offline
Micromolar User
Micromolar User
User avatar

Joined: Fri Sep 18, 2009 9:49 am
Posts: 13
Dear mswingle:

Thanks a lot for your help. I really appriciate your kindness. To help more people, I'd also like to post the question in my PM content here:

Quote:
Dear mswingle:

Thanks a lot for your reply to my post. Actually, I docked a ligand to a receptor using both vina and autodock4. For AD4, the rmsd between the dockings and reference structure is calculated automatically. While for vina, it is not.

Thus I need to find a methond to calculate rmsd in the same way to decide:the conformations from which software are closer to the reference structure.

Do you know how to calculate rmsd in pymol or other tools taking symmetry into account as AD4? Thanks.


mswingle wrote:

stonefonly,

I'm going to answer your PM here for the benefit of future generations :P

You can use autodock to do the rmsd calculations with, e.g. Vina's output coordinates.

This is kind of kludgey - maybe some one knows of an easier way?

1st, load the docked conformation's coordinates into ADT. Go to Docking-->Ligand-->Choose and select the docked ligand. Then, go to Docking-->Ligand-->Ligand Parameters. In the parameter window, check "yes" for "Specify Initial Dihedrals?" and then uncheck all of the boxes that have the word "Random" next to them (i.e. for Initial position, relative dihedral offset, and dihedrals).

Now, set up a .dpf like the following:

---------------------------------------------------------------
ligand_types C OA
fld whatever.maps.fld
map whatever.C.map
map whatever.OA.map
elecmap whatever.e.map
desolvmap whatever.d.map
move blah-docked.pdbqt
about 27.976 -1.335 15.659
rmsref blah-crystal.pdbqt
tran0 27.976 -1.335 15.659
quat0 1 0 0 0
dihe0 0 0
ls_search_freq 0
set_sw1
do_local_only 1
analysis
-------------------------------------------------

"about" should be the same as "tran0", which are the xyz coordinates from the "User specified initial position" field in ADT's ligand parameter window. "move" is your docked conformation and "rmsref" is your crystallographic reference structure. "quat0" is "1 0 0 0" and "dihe0" is a zero for each torsion in your ligand.

Alter "ligand_types" , "fld", "map" etc. to fit your ligand and receptor. I don't think you really need to specify all of the ligand types and maps unless you need scores in addition to rmsd values.

When you run this, autodock will start a local search from the specified conformation. However, since the local search probability is set to zero, this initial conformation is never changed during the "search".

Your .dlg should show the rmsd from the reference near the end of the file like so:

--------------------------------------
Sorry! Unable to perform cluster analysis, because not enough conformations were generated.


MODEL 1
USER Run = 1
USER Cluster Rank = 1
USER Number of conformations in this cluster = 1
USER
USER RMSD from reference structure = 0.373 A
USER
USER Estimated Free Energy of Binding = -8.96 kcal/mol [=(1)+(2)+(3)-(4)]
----------------------------------------


Hope this helps!


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
POWERED_BY
Translated by MaĆ«l Soucaze © 2009 phpBB.fr