Hi Vineet,
I am not sure that you are defining things as expected in the CNS
documentation. If you haven't
already, please read the CNS FAQ page with description of how to define and
include alternate
side-chains for refinement.
In any case, I summarise the steps for you-
1. Run generate.inp with your pdb file. This will make, say, files start.pdb
and start.mtf.
2. Next, run alternate.inp specifying which residues you want alternate
side-chain conformations
for. This will give you, say, alt.pdb and alt.mtf. The alt.pdb file will
contain the following
definitions: the 1st conformation as *AC1* and will list the 2nd conformer at
the end of the file
with ID *AC2*. From the coordinates, you will see that CNS has simply made a
copy of the side chains.
3. Copy the *AC2* conformations into a new pdb file since I found that the
program 'O' seems to have
trouble recognizing AC2, if AC1 exists in the same pdb file. Don't know how
Coot handles this. Read
this into O. Fix all AC2 to their correct conformation in O. Read out the
conformations.
4. Replace new AC2 conformations at the end of alt.pdb. Say, we call it
alt_new.pdb. Use this pdb
file for refinement. Throughout refinement, you need to specify the new atom
definitions in
minimize.inp and other input files.
Example:
{* select atoms in alternate conformation 1 *}
{===>} conf_1=(segid AC1 AND RESID 46);
{* select atoms in alternate conformation 2 *}
{===>} conf_2=(segid AC2 AND RESID 46);
5. Most importantly, use *alt.mtf* throughout this refinement cycle.
6. Also, make sure the occupancies for AC1 and AC2 are distributed - some folks
split occu as 0.5
and 0.5. Some split it as 0.65 and 0.35. You may have other intelligent
criteria to determine the
occupancies.
7. Remember, you need to carry out all of the above steps at every round of
refinement. So, keep
alternate side-chain addition to the very final rounds of structure
determination.
8. If it still remains a black box, switch to Refmac.
Hope that helps.
Raji
---------Included Message----------
>Hi all
>
>Sorry for a non CCP4 question.
>I m using CNS for structure refinement. The structure is having few residues
>in alternate confirmations. i can see the density for those alternate
>confirmations. i m defining those alternate confirmations in the pdb but
>while generating the topology file, the information of alternate
>confirmations is getting lost. how can i define the alternate confirmations
>in generate.inp n minimize.inp.
>
>the way i m defining the alternate confirmations in pdb :
>
>
>
>ATOM 6101 1N LYS D 93 83.029 39.489 93.510 0.50 42.90 D N
>
>ATOM 6101 2N LYS D 93 83.028 39.495 93.498 0.50 42.90 D N
>
>ATOM 6102 1CA LYS D 93 82.366 40.672 93.101 0.50 41.09 D C
>ATOM 6102 2CA LYS D 93 82.391 40.686 93.075 0.50 41.09 D C
>ATOM 6103 1CB LYS D 93 83.019 41.834 93.797 0.50 41.27 D C
>ATOM 6103 2CB LYS D 93 83.119 41.836 93.691 0.50 41.27 D C
>ATOM 6104 1CG LYS D 93 82.331 42.627 94.824 0.50 43.61 D C
>ATOM 6104 2CG LYS D 93 82.419 42.747 94.517 0.50 43.61 D C
>ATOM 6105 1CD LYS D 93 81.797 42.040 96.162 0.50 43.24 D C
>ATOM 6105 2CD LYS D 93 82.004 42.209 95.769 0.50 43.24 D C
>ATOM 6106 1CE LYS D 93 82.265 40.647 96.659 0.50 43.68 D C
>ATOM 6106 2CE LYS D 93 80.982 43.099 96.495 0.50 43.68 D C
>ATOM 6107 1NZ LYS D 93 82.884 40.390 97.999 0.50 43.36 D N
>ATOM 6107 2NZ LYS D 93 80.832 44.665 96.562 0.50 43.36 D N
>ATOM 6108 1C LYS D 93 82.478 40.825 91.578 0.50 39.28 D C
>
>ATOM 6108 2C LYS D 93 82.477 40.826 91.567 0.50 39.28 D C
>
>ATOM 6109 1O LYS D 93 82.412 41.847 91.106 0.50 38.72 D O
>
>ATOM 6109 2O LYS D 93 82.408 41.844 91.093 0.50 38.72 D O
>
>
>Thanks in advance
>
>vineet gaur
>
>
---------End of Included Message----------
