5
$\begingroup$

I am trying to convert http://crystallography.net/cod/2236556.cif to a file that I can open in Avogadro without some bonds disappearing. See this crystal structure is of tianeptine hydrochloride, yet many of the bonds are not shown when I open the CIF in Avogadro.

This is what the CIF is meant to be rendered like (shown in PyMOL):

enter image description here

while this is what I see in Avogadro (also displayed like this in Avogadro 2):

enter image description here

note how so many bonds are not shown properly. I suspect if I could convert the molecule to SDF or CML or some other 3D chemical format I'd be able to fix this issue and Avogadro will be able to render the file properly.

I have tried doing the file conversion with OpenBabel but the files generated look, when viewed in PyMOL or Jmol (which render the CIF fine), just like the CIF does in Avogadro with all the disconnected atoms.

$\endgroup$
5
  • $\begingroup$ Similar issue has been discussed here, where use of Mercury from CCDC suit is suggested. I can confirm that SDF/MOL-file generated with Mercury behaves as expected in Avogadro with proper connectivity. The problem is, Mercury is not open-source, though it can be run under Linux, so I'm not sure if it solves anything for you. $\endgroup$
    – andselisk
    Jul 16, 2017 at 4:04
  • $\begingroup$ I actually had it already installed just didn't realize it could do that. Is it possible to automatically determine bond order in it too? $\endgroup$
    – Josh Pinto
    Jul 16, 2017 at 4:07
  • $\begingroup$ I don't know for sure, it seems like MOL-file generated with Mercury doesn't have bond order info. I suspect it's more software-dependent (how it treats certain interatomic distances). $\endgroup$
    – andselisk
    Jul 16, 2017 at 4:14
  • $\begingroup$ Write it into an answer and I'll accept it. I've found another way to determine bond order automatically so I'm happy as far as that goes. $\endgroup$
    – Josh Pinto
    Jul 16, 2017 at 4:21
  • $\begingroup$ I can confirm this was actually an inadvertent bug in Open Babel 2.4.x. I'm currently fixing the behavior, which will go into the newest releases of Open Babel and Avogadro. $\endgroup$ Apr 17, 2018 at 21:03

1 Answer 1

5
$\begingroup$

It seems like OpenBabel and Avogadro sometimes don't ideally parse certain CIF-files' connectivity data. As OP mentioned, PyMOL and other crystallographically-oriented programs (I personally like Olex2, it also run under Linux and is free) do it on-the-fly:

enter image description here

As @GeoffHutchison suggested in a similar thread on binding atoms within a unit cell, Mercury does a good job at providing MOL/SDF files from CIF data. As of now, there is an installation script for Linux for Mercury 3.9. Avogadro handles such MOL file from Mercury way better:

enter image description here

2236556
  --CCDC--071617    3D                              

 56 57  0  0  0  0  0  0  0  0999 V2000
    1.3664    7.7013    4.0020 Cl     0
    6.6663    8.7266    4.4355 S      0
    6.1525    9.4153    5.5710 O      0
    7.7255    7.7726    4.6043 O      0
    4.7213    0.8967    7.6943 O      0
    2.6223    1.3499    8.2786 O      0
    2.9354    1.3046    9.0347 H      0
    7.1953    9.8372    3.3639 N      0
    7.6456    5.9591    2.4701 N      0
    8.4385    5.8198    2.0682 H      0
    7.8134    6.2148    3.3161 H      0
    5.3238    7.9653    3.5835 C      0
    4.0381    8.0880    4.1077 C      0
    3.8953    8.5011    4.9288 H      0
    2.9772    7.5765    3.3740 C      0
    3.1781    6.9816    2.1438 C      0
    2.4563    6.6559    1.6556 H      0
    4.4680    6.8774    1.6459 C      0
    4.6013    6.4768    0.8168 H      0
    5.5699    7.3547    2.3503 C      0
    6.9378    7.0993    1.7546 C      0
    6.7575    6.7565    0.8547 H      0
    7.9197    8.2523    1.5451 C      0
    8.8050    8.0470    0.4790 C      0
    8.7594    7.2477    0.0049 H      0
    9.7458    8.9915    0.1090 C      0
   10.3266    8.8270   -0.5988 H      0
    9.8106   10.1839    0.8033 C      0
   10.4293   10.8326    0.5572 H      0
    8.9631   10.4081    1.8540 C      0
    9.0175   11.2135    2.3156 H      0
    8.0176    9.4560    2.2492 C      0
    6.3410   11.0331    3.1948 C      0
    6.8195   11.7005    2.6977 H      0
    6.1061   11.3837    4.0569 H      0
    5.5420   10.7916    2.7197 H      0
    6.8948    4.6534    2.5072 C      0
    7.5354    3.9263    2.4577 H      0
    6.3296    4.5991    1.7205 H      0
    6.0338    4.4553    3.7359 C      0
    5.2898    5.0768    3.7165 H      0
    6.5567    4.6337    4.5332 H      0
    5.5123    3.0333    3.7738 C      0
    5.0259    2.8601    2.9524 H      0
    6.2680    2.4268    3.8010 H      0
    4.6053    2.7307    4.9451 C      0
    3.8347    3.3179    4.9116 H      0
    5.0816    2.9111    5.7713 H      0
    4.1326    1.2872    4.9481 C      0
    4.8942    0.7100    5.1149 H      0
    3.7873    1.0713    4.0679 H      0
    3.0575    0.9929    5.9842 C      0
    2.3293    1.6247    5.8729 H      0
    2.7039    0.1030    5.8300 H      0
    3.5713    1.0798    7.3895 C      0
    6.3200    5.1537   -1.1307 Cl     0
  1 15  1
  2  3  1
  2  4  1
  2  8  1
  2 12  1
  5 55  1
  6  7  1
  6 55  1
  8 32  1
  8 33  1
  9 10  1
  9 11  1
  9 21  1
  9 37  1
 12 13  1
 12 20  1
 13 14  1
 13 15  1
 15 16  1
 16 17  1
 16 18  1
 18 19  1
 18 20  1
 20 21  1
 21 22  1
 21 23  1
 23 24  1
 23 32  1
 24 25  1
 24 26  1
 26 27  1
 26 28  1
 28 29  1
 28 30  1
 30 31  1
 30 32  1
 33 34  1
 33 35  1
 33 36  1
 37 38  1
 37 39  1
 37 40  1
 40 41  1
 40 42  1
 40 43  1
 43 44  1
 43 45  1
 43 46  1
 46 47  1
 46 48  1
 46 49  1
 49 50  1
 49 51  1
 49 52  1
 52 53  1
 52 54  1
 52 55  1
M  END
$$$$
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.