smina / Discussion / Help and Feedback: different docking poses from different input type

Thomas Evangelidis - 2019-06-03

I am really troubled by the output of Smina. One of the its greatest advantages over Vina was that there is no need to convert to PDBQT, Smina can read in a multi-mol SDF file for example and dock every compound onto the receptor. Below I demonstrate that:

the docking poses and energies obtain from SDF input are different than from PDBQT input although I use the same --seed and parameter values.

the protonations of the output files are also different, although I use '--addH off' to prevent the code from protonating.

Please look at the attached tarball for input and commands of an example case. I also include the output poses for your convenience (_posesfromSDF.sdf, _posesfromPDBQT.sdf) If this is a bug in the code then I believe it warrants immediate attention.

Best,
Thomas

LEM00018316.tar.gz
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- David Koes - 2019-06-04
  
  Hi Thomas,
  
  There’s a few things going on here. First, since you are docking the individual pdbqts with separate invocations of smina, only the first set of output poses is expected to be the same as the all-at-once sdf docking, since the random number generator sequence will be different for the later poses.
  
  For the first conformer docked (LEM00018316_stereo1_ion1_tau1), although many of the output poses are the same or similar, some are still different. This doesn’t appear to be due to protonation, but due to openbabel not roundtripping aromatic bonds. If you convert the sdf to pdbqt to sdf to pdbqt, you’ll see the aromaticity differs between the two pdbqts. If you dock the second pdbqt, you get the same output as with the sdf.
  
  In addition to this, keep in mind that sdf has 4 decimal places while pdbqt has 3. In some cases even these small difference can result it large differences in the final solution.
  
  David Koes
  
  Assistant Professor
  Computational & Systems Biology
  University of Pittsburgh
  
  On Jun 3, 2019, at 6:36 AM, Thomas Evangelidis tevang@users.sourceforge.netamp#116;amp#101;amp#118;amp#97;amp#110;amp#103;amp#64;amp#117;amp#115;amp#101;amp#114;amp#115;amp#46;amp#115;amp#111;amp#117;amp#114;amp#99;amp#101;amp#102;amp#111;amp#114;amp#103;amp#101;amp#46;amp#110;amp#101;amp#116; wrote:
  
  I am really troubled by the output of Smina. One of the its greatest advantages over Vina was that there is no need to convert to PDBQT, Smina can read in a multi-mol SDF file for example and dock every compound onto the receptor. Below I demonstrate that:
  
  the docking poses and energies obtain from SDF input are different than from PDBQT input although I use the same --seed and parameter values.
  
  the protonations of the output files are also different, although I use '--addH off' to prevent the code from protonating.
  
  Please look at the attached tarball for input and commands of an example case. I also include the output poses for your convenience (_posesfromSDF.sdf, _posesfromPDBQT.sdf) If this is a bug in the code then I believe it warrants immediate attention.
  
  Best,
  Thomas
  
  different docking poses from different input typehttps://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fdiscussion%2Fhelp%2Fthread%2F158f4e1005%2F%3Flimit%3D25%23c776&data=02%7C01%7Cdkoes%40pitt.edu%7Ced4391fc94e54b14020508d6e80f672f%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636951550182126726&sdata=O96NQDPEN28OVwyBsGVv3Xl5XjpSwrH4pqaLtpuDyb8%3D&reserved=0
  
  Sent from sourceforge.nethttp://sourceforge.net because dkoes@pitthttps://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fu%2Fpitt%2F&data=02%7C01%7Cdkoes%40pitt.edu%7Ced4391fc94e54b14020508d6e80f672f%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636951550182136721&sdata=0%2FS%2B5FDDLPHuj21sZ%2FvnqE28NME0fFy1vz5BZtjWP4o%3D&reserved=0.edu is subscribed to https://sourceforge.net/p/smina/discussion/help/https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fdiscussion%2Fhelp%2F&data=02%7C01%7Cdkoes%40pitt.edu%7Ced4391fc94e54b14020508d6e80f672f%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636951550182136721&sdata=%2Fx%2FRmhd9fNnEDmwPpWSDC7L4B6jYx%2Br0XL1Ta%2BR7mME%3D&reserved=0
  
  To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/smina/admin/discussion/forums.https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fadmin%2Fdiscussion%2Fforums.&data=02%7C01%7Cdkoes%40pitt.edu%7Ced4391fc94e54b14020508d6e80f672f%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636951550182146716&sdata=1PhHNpGdHvHBPKw04hcfjZzzbbG1GTXgPUsgCugB0WE%3D&reserved=0 Or, if this is a mailing list, you can unsubscribe from the mailing list.
  
  alternate
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
  - Thomas Evangelidis - 2019-06-06
    
    Hi David,
    
    Thank you for the clarifications. What is the recommended way to go? Can I safely use the multi-molecule SDF as input in just one dockign run? If not, how to safely convert SDF to PDBQT, with 'prepare_ligand4.py" from AutoDock Tools? I currently do that with OpenBabel.
    
    Best,
    Thomas
    
    If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
    - David Koes - 2019-06-06
      
      I think it’s perfectly safe to use the sdf.
      
      David Koes
      
      Assistant Professor
      Computational & Systems Biology
      University of Pittsburgh
      
      On Jun 6, 2019, at 8:35 AM, Thomas Evangelidis tevang@users.sourceforge.netamp#116;amp#101;amp#118;amp#97;amp#110;amp#103;amp#64;amp#117;amp#115;amp#101;amp#114;amp#115;amp#46;amp#115;amp#111;amp#117;amp#114;amp#99;amp#101;amp#102;amp#111;amp#114;amp#103;amp#101;amp#46;amp#110;amp#101;amp#116; wrote:
      
      Hi David,
      
      Thank you for the clarifications. What is the recommended way to go? Can I safely use the multi-molecule SDF as input in just one dockign run? If not, how to safely convert SDF to PDBQT, with 'prepare_ligand4.py" from AutoDock Tools? I currently do that with OpenBabel.
      
      Best,
      Thomas
      
      different docking poses from different input typehttps://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fdiscussion%2Fhelp%2Fthread%2F158f4e1005%2F%3Flimit%3D25%23c776%2F969b%2F9acb&data=02%7C01%7Cdkoes%40pitt.edu%7Cc96fbcb5453846e3bc6d08d6ea7b87bc%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636954213608116149&sdata=EpoY1UUIO1RydCO6R4EWNFynJBV7sXJmiJWrUDOprpk%3D&reserved=0
      
      Sent from sourceforge.nethttp://sourceforge.net because dkoes@pitthttps://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fu%2Fpitt%2F&data=02%7C01%7Cdkoes%40pitt.edu%7Cc96fbcb5453846e3bc6d08d6ea7b87bc%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636954213608116149&sdata=7HnQaLClZ%2FJTcfc1g%2FfjWr7v6vTfajYK2DmmrCJwlJA%3D&reserved=0.edu is subscribed to https://sourceforge.net/p/smina/discussion/help/https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fdiscussion%2Fhelp%2F&data=02%7C01%7Cdkoes%40pitt.edu%7Cc96fbcb5453846e3bc6d08d6ea7b87bc%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636954213608126145&sdata=pva2kzyomGzOfczx%2BUqUbOE0RKZgSrjNwttFxLQmkQA%3D&reserved=0
      
      To unsubscribe from further messages, a project admin can change settings at https://sourceforge.net/p/smina/admin/discussion/forums.https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsourceforge.net%2Fp%2Fsmina%2Fadmin%2Fdiscussion%2Fforums.&data=02%7C01%7Cdkoes%40pitt.edu%7Cc96fbcb5453846e3bc6d08d6ea7b87bc%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1%7C0%7C636954213608126145&sdata=sYneAOjRZ5EpzpXRp1gjjHxGLFc4jeD03I5%2FzCi6vZA%3D&reserved=0 Or, if this is a mailing list, you can unsubscribe from the mailing list.
      
      alternate
      
      If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

different docking poses from different input type

Scoring and Minimization with AutoDock Vina

Forums

Help

different docking poses from different input type

different docking poses from different input type

Scoring and Minimization with AutoDock Vina

Forums

Help

different docking poses from different input type document.SUBSCRIPTION_OPTIONS = { "thing": "topic", "subscribed": false, "url": "subscribe", "icon": { "css": "fa fa-envelope-o" } };

different docking poses from different input type