On Saturday, 20 April 2013, Jinzhong Lin wrote:
> Dear Coot users,
>
> I have been refining the crystal structure of a large assembly which has
> half a million of atoms. In the PDB, when atom number exceeds 99999, it
> starts to use letters such as A0000. Everything was fine for phenix and
> coot to deal with such a structure until recently I tried to refine the
> TLS parameters. The TLS refinement puts ANISOU record below each ATOM
> line in the PDB file and when the atom number has 5 digits it
> concatenates with ANISOU giving something like:
>
> ATOM 99999 O2* Cr A 453 41.217 18.570 115.721 1.00107.41
> 16S1 O
> ANISOU99999 O2* Cr A 453 18694 11085 11032 -1586 395 -1101
> 16S1 O
> ATOM A0000 C1* Cr A 453 42.986 17.740 114.321 1.00106.92
> 16S1 C
> ANISOUA0000 C1* Cr A 453 18592 11040 10992 -1526 256 -1044
> 16S1 C
Lovely. Not.
I really disapprove of the idea of writing out ANISOU records to
describe TLS refinement. Notwithstanding the fact that the PDB
recommends this, I strongly suggest that you tell refmac _not_
to write out ANISOU records. The TLS model is adequately
described in the header, the extra ANISOU records are a less
good way of describing the same thing.
Ethan
> When I try to open the PDB file in coot, it gives an error:
>
> ERROR 14 READ: Unmatch in different records for the same atom.
> LINE #200536
> ANISOUA0000 C1* Cr A 453 18592 11040 10992 -1526 256
> -1044 16S1 C
>
> It looks like coot does not recognize ANISOU followed by a letter
> without space in between. (ANISOU99999 is fine)
>
> I am not sure if it is a bug of coot or it is inherited from mmdb. I
> would appreciate it very much if anyone can help me out or give me a clue.
>
> The TLS refinement gives me significant improvement in both map and
> RFree, It's a shame I can not use it.
>
> Thanks.
>
>
> Best Regards,
> lithlin
>
--
Tradition is not the worship of ashes, but the preservation of fire.
- Gustav Mahler
