[Chimera-users] morph interpolation
souro at ebi.ac.uk
Thu Jul 20 02:07:14 PDT 2017
My apologies for the delayed response. I can now confirm that the movement of the hydrogen has been fixed. Thank you so much for going through this and helping me identify the errors! :-)
> On 12 Jul 2017, at 00:36, Elaine Meng <meng at cgl.ucsf.edu> wrote:
> HI Souro,
> Morphing requires (1) pairing atoms in initial and final structures (2) dividing up the atoms in sets that move together in 3D, which tries to make atoms in the same residue move together.
> (1) The atom pairing is described in the same link I sent earlier today. In ligands, requires same names.
> (2) The H01 on H1082 moves around because it is in the same residue as the moving PO4. So if you didn’t want it to move around, you should edit the file so that it’s in the correct residue (H1082).
> However, the incorrect residue of H1082 may have fortuiously caused the ligand to be included, because ligands must be bonded to the protein to be included in the morph, as explained in the link above. If you fix it, you’d have to add some fake bonds to include the ligand, which could be hidden later.
> You would need to do the following in both input structures:
> (A) edit the PDB file to put H01 in the correct residue if you don’t want it to move in the morph (change its residue from PO4 to HIS 1082) and change its name to HD1 because it is on ND1 and change that line from HETATM to ATOM
> (B) fix the other problems in H1082 if you don’t want it to look crazy (delete all the CONECT lines at the end of the file, change name of H1082 atom HD1 to HE2 because it is on NE2)
> (C) add a bond between the ligand and the protein because that’s required to make the ligand morphing work in Chimera, must be between the same two atoms in both structures. In each structure, you could do that by Ctrl-clicking the two atoms and using command “bond sel”, or commands like
> bond #0 at c1':1082 at ne2
> bond #1 at c1':1082 at ne2
> Then in the morph you can hide any bonds you don’t want to see by Ctrl-clicking to select it, using the Selection Inspector (green magnifying glass icon on lower left) to change it to displayed “false” instead of “if atoms shown”.
> However, the PO4 is still disconnected. Really it should be made part of residue T1. I tried to fix all these problems for you but I ran out of patience when I encountered this problem.… it is your project, after all.
> Because I’m nice, :-) (and I already spent the time anyway) I attach files fix2.pdb and fix3.pdb that if you open, then run the above bonding commands, and then create the morph, seem to have all the problems fixed except not including the PO4 in the morph. I leave any further fixing to you.
> Stepa (A) and (B) above are to fix errors in the files. Even if some program caused the problems in the files, they are still errors. (C) is a weird Chimera-specific thing to make morphing include the ligand residue(s).
> I hope this helps,
> Elaine C. Meng, Ph.D.
> UCSF Chimera(X) team
> Department of Pharmaceutical Chemistry
> University of California, San Francisco
>> On Jul 11, 2017, at 2:04 PM, Souroprobho Chowdhury <souro at ebi.ac.uk> wrote:
>> Hi Elaine,
>> Thank you for getting back with your detailed response. The 6574 is part of the PO4 residue because that is how Maestro ‘preserves’ the information that the atom has originally come from a phopsphate? I am using Maestro to do the modelling, essentially breaking a X—H bond and making a Y—H bond with the same hydrogen.
>> Essentially we are trying to make ‘movies’ of 3D enzyme mechanisms. This means that morph interpolate should be able to know that atom X from residue N goes to position Y of residue M. Am I missing something obvious?
>> Souro Chowdhury
>> Researcher, Thornton Group
>> European Bioinformatics Institute, EMBL-EBI
>> Cambridge, United Kingdom
>>> On 11 Jul 2017, at 19:09, Elaine Meng <meng at cgl.ucsf.edu> wrote:
>>> Hi Souro,
>>> Your two input files I2.pdb and I3.pdb have severe errors. Looking at I2.pdb with a text editor, I see that the hydrogen that is supposedly on the histidine is actually part of the PO4 residue according to that file, and the connectivity of the histidine-1082 sidechain is badly messed up. For example, hydrogen HD1 with serial number 3865 is bonded to two other atoms (according to the CONECT section at the bottom of the file) . See image attached with atoms labeled by serial number and “ligand” selected, so that you can see that hydrogen 6574 is part of the ligand residue PO4 even though the CONECT lines have it bonded to the histidine.
>>> When I tried the morphing, it seemed to me that the movement of O3 was reasonable, however (just goes between starting and ending positions).
>>> Initially I was going to answer that Chimera's morphing was developed mainly for the biopolymer parts (protein and nucleic chains) and I wouldn’t necessarily expect good performance on ligand conformations. However, when I saw how badly messed up the input files are, now I think it might work a lot better if those connectivity issues (and putting the H in the histidine residue instead of the PO4 residue) were fixed.
>>> I don’t use Pymol, so I didn’t look at that script. If morphing works in pymol it may be because it is ignoring the connectivity information (CONECT lines at the bottom) in your PDB files.
>>> I hope this clarifies the situation. Best,
>>> Elaine C. Meng, Ph.D.
>>> UCSF Chimera(X) team
>>> Department of Pharmaceutical Chemistry
>>> University of California, San Francisco
>>>> On Jul 11, 2017, at 6:25 AM, Souroprobho Chowdhury <souro at ebi.ac.uk> wrote:
>>>> I am trying to use morph interpolate on two (for this post) PDBs (I2.pdb and I3.pdb, attached) but something peculiar is happening. The Hydrogen on the HIS 1082 (N1) does a ‘cork-screw’ like movement. Essentially, instead of interpolating directly from position A to B, it is taking a more round-about angle. A similar movement is observed for the O3 in the T1 O ligand.
>>>> This problem has come up recurrently in my work and I am yet to find a solution. I have used most of the options that apply on morph interpolate that are mentioned on the chimera website without much improvement.
>>>> I am using Maestro by Schrodinger to model my structures.
>>>> Please find attached I2.pdb , I3.pdb and mov_all.pdb. mov_all.pdb is the trajectory file, works on Pymol (I am attaching a pymol script called auto-script.pml as well for selection of the correct residues, and a reasonable representation). The chimera interpolation script is also attached.
>>>> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
>>>> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
>> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Chimera-users