[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Modelling Dimers



Hi Michael

I think a more elegant solution is to treat the chains separately.  If you
want them to stay the same you can apply restraints.  Thus the alignment
file might look like this

>P1;candida_dimer_1st
sequence:candida_dimer_1st::::
----------------------M
KIVLVL------YDAGKHAADE-EKLYGCTENKLGIANWLKDQGHELITT
SDKEGGNSVLDQHIPDADIIITTPFHPAYITKERIDKAKKLKLVVVAGVG
SDHIDLDYINQTGKKISVLEVTGSNVVSVAEHVVMTMLVLVRNFVPAHEQ
IINHDWEVAAIAKDAYDIEGKTIATIGAGRIGYRVLERLVPFNPKELLYY
DYQALPKDAEEKVGARRVENIEELVAQADIVTVNAPLHAGTKGLINKELL
SKFKKGAWLVNTARGAICVAEDVAAALESGQLR
GYGGDVWFPQTAPKDHPWRDMRNKYGAGNAMTPHYSGTTLDAQTRYAQG
TKNILESFFTGKFDYRPQDIILLNGEYVTKAYGKHDKK-------
/
----------------------M
KIVLVL------YDAGKHAADE-EKLYGCTENKLGIANWLKDQGHELITT
SDKEGGNSVLDQHIPDADIIITTPFHPAYITKERIDKAKKLKLVVVAGVG
SDHIDLDYINQTGKKISVLEVTGSNVVSVAEHVVMTMLVLVRNFVPAHEQ
IINHDWEVAAIAKDAYDIEGKTIATIGAGRIGYRVLERLVPFNPKELLYY
DYQALPKDAEEKVGARRVENIEELVAQADIVTVNAPLHAGTKGLINKELL
SKFKKGAWLVNTARGAICVAEDVAAALESGQLR
GYGGDVWFPQTAPKDHPWRDMRNKYGAGNAMTPHYSGTTLDAQTRYAQG
TKNILESFFTGKFDYRPQDIILLNGEYVTKAYGKHDKK-------
*
>P1;2nad_doubleA_noHET
structureX:2nad_doubleA_noHET:1:A:391:B
AKVLCVLYDDPVDGYPKTYARDD
LPKIDHYPGGQTLPTPKAIDFTPGQLLGSVSGELGLRKYLESNGHTLVVT
SDKDGPDSVFERELVDADVVISQPFWPAYLTPERIAKAKNLKLALTAGIG
SDHVDLQSAID--RNVTVAEVTYCNSISVAEHVVMMILSLVRNYLPSHEW
ARKGGWNIADCVSHAYDLEAMHVGTVAAGRIGLAVLRRLAPFD-VHLHYT
DRHRLPESVEKELNLTWHATREDMYPVCDVVTLNCPLHPETEHMINDETL
KLFKRGAYIVNTARGKLCDRDAVARALESGRLA
GYAGDVWFPQPAPKDHPWRTMP-----YNGMTPHISGTTLTAQARYAAG
TREILECFFEGR-PIRDEYLIVQGGALAGTGAHSYSKGNATGGSE
/
AKVLCVLYDDPVDGYPKTYARDD
LPKIDHYPGGQTLPTPKAIDFTPGQLLGSVSGELGLRKYLESNGHTLVVT
SDKDGPDSVFERELVDADVVISQPFWPAYLTPERIAKAKNLKLALTAGIG
SDHVDLQSAID--RNVTVAEVTYCNSISVAEHVVMMILSLVRNYLPSHEW
ARKGGWNIADCVSHAYDLEAMHVGTVAAGRIGLAVLRRLAPFD-VHLHYT
DRHRLPESVEKELNLTWHATREDMYPVCDVVTLNCPLHPETEHMINDETL
KLFKRGAYIVNTARGKLCDRDAVARALESGRLA
GYAGDVWFPQPAPKDHPWRTMP-----YNGMTPHISGTTLTAQARYAAG
TREILECFFEGR-PIRDEYLIVQGGALAGTGAHSYSKGNATGGSE

and at the end of your top file you put something like this

SUBROUTINE ROUTINE = 'special_restraints'

# Try to put symmetry restraints here
# This is called from __homcsr after other restraints set up

       SET RES_TYPES = 'ALL'
       SET ATOM_TYPES = 'ALL'
       SET SELECTION_STATUS = 'INITIALIZE'
       SET SELECTION_SEARCH = 'SEGMENT'

       SET SYMMETRY_WEIGHT = 0.5
       PICK_ATOMS PICK_ATOMS_SET = 2, SELECTION_SEGMENT = '1:' '364:'
       PICK_ATOMS PICK_ATOMS_SET = 3, SELECTION_SEGMENT = '365:' '728:'
       DEFINE_SYMMETRY ADD_SYMMETRY = on off

  RETURN
END_SUBROUTINE

Residues 1-364 are the first chain and 365-728 the second chain of the
output homodimer.

Good luck

Daniel

+-------------------------------------------------------------------------+
|                       Dr Daniel John Rigden                             |
| CENARGEN/EMBRAPA                 | e-mail:    |
| Parque Estacao Biologica         | http://www.cenargen.embrapa.br       |
| PqEB - Final - Av. W3 Norte      |  Phone:  +55 (61)448-4741            |
| 70770-900, Brasilia-D.F.-BRAZIL  |  Fax:    +55 (61)340-3658            |
+-------------------------------------------------------------------------+



On Thu, 25 Jul 2002, Michael Buck wrote:

> I have a question about modelling dimers.  What I am trying to do is model
> a homodimer.  I have tried to few things that might work, but I am not
> really sure.  What I did:
>          1.  Edited the pdb file for the template, making it all one subunit
>          2.  My alignment file has the 60 amino acids from the first dimer
> then chain of glycines then next 60 amino acids from the second dimer.
>          3.  After I model, remove the chain of glycines and add back the B
> domain letter to the pdb file.
>
> The models look good.  What does the group think?  Will these models be
> accurate?  Will the restraints between the two dimers be included in the
> models?  Can I edit the restraints to remove the bad restraints located at
> the end of the first dimer and beginning of the second?
>
>
>
>
>
> Below is the alignment: I edited the 1a0a template change B domain to A and
> adjusting the amino acid numbers.
>  >P1;1a0aSP
> structureX:1a0aSP:1:A:112:A:Pho4:yeast: 2.00:-1.00
> MKRESHKHAEQARRNRLAVALHELASLIPAEWKQQN---VSAAPSKATTVEAACRYIRHLQQNGST--------------MKRESHKHAEQARRNRLAVALHELASLIPAEWKQQN---VSAAPSKATTV*
>  >P1;1hlo
> sequence:1hloModel: : : ::Max:Human: 2.00:-1.00
> DKRAHHNALERKRRDHIKDSFHSLRDSVP--SLQGE------KASRAQILDKATEYIQYMGGGGGGGGGGGGGGGGGGGGDKRAHHNALERKRRDHIKDSFHSLRDSVP--SLQGE------KASRAQIL*
>
>
>
>
>
>
>
> ************************************************
> Michael Buck
> NCSU Genetics
> 
> Phone  (919)515-5759
> Fax        (919)515-3355
> http://www4.ncsu.edu/~mjbuck
> *************************************************
>
>