Re: [ccp4bb]: cif files of unknown ligands

["Eleanor J. Dodson" <ccp4@ysbl.york.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

sameeta bilgrami wrote:
> 
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> dear all,
>      hello,
> i have a ligand in my protein structure which i am
> trying to refine... i made the cif file using sketcher
> ......but refmac does not seem to run ...however if i
> put the ligand far enough from the protein molecule
> (without it having any contacts), refmac does not seem
> to have any problems...i tried including the link
> statement in header of the pdb file but that also did
> not work....there is also a cif file made by refmac
> itself ...if i use that file the program seems to run
> but the geometry of the protein becomes bad and so
> does the density....i compared the cif file made by
> the sketcher and that made by refmac ...the one made
> by refmac seems also to have link information in it
> ..... so how do we include the correct link information
> in the cif file ...is there a way to do it by means of
> monomer library sketcher ....thanx in advance for the help...sameeta
> 


 Presumably the cif file for the ligand is correct? 

The best way to check is to use the GUI.

Once you have the ligand descriotion run refinement with REVIEW
RESTAINTS option,
and reading in the ligand cif decription.
I usually turn on MINOTOR MANY ( see under monitoring folder.

 Then you need to find the appropriate analysis in the log file.. 


 You will find various sorts of information after the (incorrect)
message..
WARNING: Chain_list is not correct. Program changed it.
           Different chains have the same chain id.
  Number of atoms    :    4655
  Number of residues :     673
  Number of chains   :       9
  I am reading library. Please wait.
                mon_lib.cif
  WARNING : description of link:HIS-HEM  not found in the dictionary.
            link will be created with bond_lenth =   1.855
  WARNING : link:HIS-HEM  is found dist =     2.155 ideal_dist=    
1.855
            ch:BB res: 235  HIS      at:NE2 .->Bb res: 291  HEM     
at:FE  .
  WARNING : link:HIS-HEM  is found dist =     2.339 ideal_dist=    
1.855
            ch:DD res: 635  HIS      at:NE2 .->Dd res: 691  HEM     
at:FE  .


 Here the program has found links betwen the HIS and Fe of the HEM, and
the pdb file output has LINK records describing these.

The output cif file has also added LINK information.

PDB output:

HEADER    ----                                    XX-XXX-XX  
xxxx              
COMPND   
---                                                                   
LINK        NE2  HIS B 235        1.855   FE   HEM B 291               
HIS-HEM
LINK        NE2  HIS D 635        1.855   FE   HEM D 691               
HIS-HEM
CRYST1   97.830   99.500   66.170  90.00  90.00  90.00 P
1                      
.....

libput.cif stuff

# ---   LIST OF LINKS ---
#
data_link_list
loop_
_chem_link.id
_chem_link.comp_id_1
_chem_link.mod_id_1
_chem_link.group_comp_1
_chem_link.comp_id_2
_chem_link.mod_id_2
_chem_link.group_comp_2
_chem_link.name
HIS-HEM  HIS      .        .        HEM      .        .
 bond_HIS-NE2_=_HEM-FE
#
# --- DESCRIPTION OF MONOMERS ---
.....
# --- DESCRIPTION OF LINKS ---
#
data_link_HIS-HEM
#
loop_
_chem_link_bond.link_id
_chem_link_bond.atom_1_comp_id
_chem_link_bond.atom_id_1
_chem_link_bond.atom_2_comp_id
_chem_link_bond.atom_id_2
_chem_link_bond.type
_chem_link_bond.value_dist
_chem_link_bond.value_dist_esd
 HIS-HEM  1 NE2     2 FE        .           1.855    0.020
# ------------------------------------------------------


 In fact I reset this distance to 2.000 and the esd to 0.04 - but you
are free to edit it.



 But you need to be careful here - if there are bad contacts the cif
file will try to describe them, and you will get meaningless links which
must be removed.

Check sketcher has assigned the planes correctly - the bond orders etc
etc..

You will need to edit the output cif file to give a correct description
of your ligand and its relationship to the protein.

 Once that is done you can run "restrained refinement" , and bad VDW
clashes should be corrected fairly reliably. But remeber - refinement
programs do what they are told, so if there is a false LINK record the
program will struggle to satisfy it!

 Good luck Eleanor



> =====
> Sameeta Bilgrami
> Research Scolar
> Deptt. of biophysics
> All India Instt. of Medical Scs.
> India
> 
> ________________________________________________________________________
> Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!!
>        visit http://in.autos.yahoo.com