Thank you all and certainly seems llike now I am going to some right direction. 

I have read some discussion part (page 14) of the paper Maia sent, as stated below, the BSA ( value for my dimer interfaces are ~1000 (as predicted by PISA) which according to the Krissinel paper, is biological relevant. Moreover, the hexameric structure is reported to exist in the same specie on which I am working on. 



It has been concluded in a number of studies [20, 68, 69, 70] that BSA larger than 600-850°A2 indicates a biologically relevant interface. A lower figure of 400 °A2 was found in Ref. [9] and then used in the Protein Quaternary Structure (PQS) server [5]. The minimal BSA of potentially stable crystal dimers in our dataset is found to be 390 °A2 (PDB entry 1SDX [67]), which agrees with the
literature data. However, it follows from Figs. 3B,5A and above considerations that unspecific interactions may prevail at ABSA · 3000°A2, causing substantial changes to the original complexes, and, therefore, dimeric structures with low ABSA may be misrepresented by crystals.



On Thu, May 20, 2010 at 1:29 AM, Maia Cherney <chern@ualberta.ca> wrote:
In his latest paper


E. Krissinel (2009). /Crystal contacts as nature's docking solutions/. J. Comp. Chem., in press; published on-line 6 May 2009; DOI 10.1002/jcc.21303

Krissinel wrote that DGdiss error is 5kcal/mol. I think that DG~5kcal/mol is a gray zone. Then he compares docking results with actual structures, a lot of failures! Is your protein exactly the same as documented or from a different species? My protein has three different oligomeric states from three different species, and the monomers from  different species superpose well.

Maia

humayun scherrif wrote:

Thank you all for the replies.
   * The protein itself makes hexamer which is well documented and
     proved structural evidence from other cytoplasmic domains ( my
     structure is also a domain).     * I have run PISA, but the online PISA server didnt give me output
     like standalone PISA in CCP4 (result is mentioned below). Online
     PISA results show that "there are not significant dimer
     interfaces and thus the trimer structure is because of only
     crystal packing result"
   * For homology modeling I didnt get any proper homologs which have
     hexameric assembly (I@ Bryn: I cant send you PDB id since its
     not submitted yet)

 Analysis of protein interfaces suggests that the following  quaternary structures are stable in solution (I wonder the DGdiss is positive value, is it significant to make Hexamer assembly because I couldnt find any help to find out about the allowed values)

 ----.-----.---------------------------------------.---------------
 Set |  No | Size  Id      ASA       BSA    DGdiss | Formula
 ----+-----+---------------------------------------+---------------
  1 |   1 |   6    0   19917.7    5536.3      3.8 |     A(2)B(2)C(2)
 ----+-----+---------------------------------------+---------------
  2 |   2 |   3    1   10722.9    2004.1      6.2 |      ABC
 ----+-----+---------------------------------------+---------------
  3 |   3 |   4    2   14004.2    3014.9      0.5 |      A(2)B(2)
    |   4 |   1    3    4217.5       0.0         -0.0 |      A
 ----+-----+---------------------------------------+---------------
  4 |   5 |   2    4    7506.2    1003.3      7.0 |        AB
    |   6 |   1    3    4217.5       0.0        -0.0 |        A
 ----+-----+---------------------------------------+---------------
  5 |   7 |   2    5    7443.8    1000.8      6.8 |      AB
    |   8 |   1    6    4282.4       0.0     -0.0 |         A
 ----+-----+---------------------------------------+---------------
  6 |   9 |   2    7    7556.5    1008.3      2.0 |      A(2)
    |  10 |   1    8    4227.1       0.0     -0.0 |        A
    |  11 |   1    3    4217.5       0.0     -0.0 |        A
 ----'-----'---------------------------------------'---------------


Waiting for your reply
Thanks

H




On Wed, May 19, 2010 at 4:41 PM, Robert Brynmor Fenwick <robert.fenwick@irbbarcelona.org <mailto:robert.fenwick@irbbarcelona.org>> wrote:

   Also, if you would like to try homology modelling then that could
   work. However you would need a couple of hexamer strucutres to
   start with. It would probably take some tinkering with current
   tools. I would probably use an MD approach to solve this problem.
   Sorry I don't have a quick fix this is not my current area of
   expertise.
   Bryn

   Sent from my iPod

   On 19/05/2010, at 09:22, humayun scherrif <hum.one@gmail.com
   <mailto:hum.one@gmail.com>> wrote:


   Thank you Bryn for your reply, But I have already tried all
   possible symmetries that it generates, but it does not provide a
   proper hexameric assembly. Does it mean this is due to problems
   in crystal packing ?
   Is there any alternative way to generate or by homology, which
   server could be suitable ?

   Regards

   H


   On Wed, May 19, 2010 at 4:02 PM, Robert Brynmor Fenwick
   <robert.fenwick@irbbarcelona.org
   <mailto:robert.fenwick@irbbarcelona.org>> wrote:

       There is a symmetry command that will build the crystal
       symmetry from
       the pdb header you could then delete the irrelevent molecules
       to leave
       the six that you want.

       Bryn

       If you have trouble with this I can hunt down the commands in
       my labbook


       > _______________________________________________
       > PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net
       <mailto:PyMOL-users@lists.sourceforge.net>)
       <mailto:users@lists.sourceforge.net>







------------------------------------------------------------------------

------------------------------------------------------------------------------

 ------------------------------------------------------------------------


_______________________________________________
PyMOL-users mailing list (PyMOL-users@lists.sourceforge.net)
Info Page: https://lists.sourceforge.net/lists/listinfo/pymol-users
Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net



--
Best Regards,

Humayun Sharif
Research Assistant
Protein Structure and Function Laboratory
Gwangju Institute Of Science & Technology,
Gwangju, 500-712, Republic of Korea
Tel (Res) :+82-62-718-4176
Tel (Lab) : +82-62-970-2549
Email: hum.one@gmail.com